In this paper, we critically examine the current "humanoid hype" in robotics, questioning its alignment with responsible robotics principles. While technical challenges drive internal fascination, the pervasive public image of humanoids demands deeper HRI engagement. We explore how responsible robotics concepts, such as privacy, dignity, and trust, are uniquely challenged or overlooked in the pursuit of anthropomorphic robot forms. By dissecting this hype, and mapping the main findings of the recently-published Roadmap for Responsible Robotics to the humanoids field, we aim to move beyond technical form-factor obsessions to understand the true societal implications and identify potential blind spots for the HRI community.

Recent learning-based visual localization methods use global descriptors to disambiguate visually similar places, but existing approaches often derive these descriptors from geometric cues alone (e.g., covisibility graphs), limiting their discriminative power and reducing robustness in the presence of noisy geometric constraints. We propose an aggregator module that learns global descriptors consistent with both geometrical structure and visual similarity, ensuring that images are close in descriptor space only when they are visually similar and spatially connected. This corrects erroneous associations caused by unreliable overlap scores. Using a batch-mining strategy based solely on the overlap scores and a modified contrastive loss, our method trains without manual place labels and generalizes across diverse environments. Experiments on challenging benchmarks show substantial localization gains in large-scale environments while preserving computational and memory efficiency. Code is available at github.com/sontung/robust_scr.

Visual Place Recognition (VPR) enables coarse localization by comparing query images to a reference database of geo-tagged images. Recent breakthroughs in deep learning architectures and training regimes have led to methods with improved robustness to factors like environment appearance change, but with the downside that the required training and/or matching compute scales with the number of distinct environmental conditions encountered. Here, we propose Hyperdimensional One Place Signatures (HOPS) to simultaneously improve the performance, compute and scalability of these state-of-the-art approaches by fusing the descriptors from multiple reference sets captured under different conditions. HOPS scales to any number of environmental conditions by leveraging the Hyperdimensional Computing framework. Extensive evaluations demonstrate that our approach is highly generalizable and consistently improves recall performance across all evaluated VPR methods and datasets by large margins. Arbitrarily fusing reference images without compute penalty enables numerous other useful possibilities, three of which we demonstrate here: descriptor dimensionality reduction with no performance penalty, stacking synthetic images, and coarse localization to an entire traverse or environmental section.

The Robot Operating System (ROS) has become the de facto standard middleware in robotics, widely adopted across domains ranging from education to industrial applications. The RoboStack distribution, a conda-based packaging system for ROS, has extended ROS's accessibility by facilitating installation across all major operating systems and architectures, integrating seamlessly with scientific tools such as PyTorch and Open3D. This paper presents ROS2WASM, a novel integration of RoboStack with WebAssembly, enabling the execution of ROS 2 and its associated software directly within web browsers, without requiring local installations. ROS2WASM significantly enhances the reproducibility and shareability of research, lowers barriers to robotics education, and leverages WebAssembly's robust security framework to protect against malicious code. We detail our methodology for cross-compiling ROS 2 packages into WebAssembly, the development of a specialized middleware for ROS 2 communication within browsers, and the implementation of www.ros2wasm.dev, a web platform enabling users to interact with ROS 2 environments. Additionally, we extend support to the Robotics Toolbox for Python and adapt its Swift simulator for browser compatibility. Our work paves the way for unprecedented accessibility in robotics, offering scalable, secure, and reproducible environments that have the potential to transform educational and research paradigms.

Visual Simultaneous Localization and Mapping (VSLAM) research faces significant challenges due to fragmented toolchains, complex system configurations, and inconsistent evaluation methodologies. To address these issues, we present VSLAM-LAB, a unified framework designed to streamline the development, evaluation, and deployment of VSLAM systems. VSLAM-LAB simplifies the entire workflow by enabling seamless compilation and configuration of VSLAM algorithms, automated dataset downloading and preprocessing, and standardized experiment design, execution, and evaluation. All of these features are accessible through a single command-line interface. The framework supports a wide range of VSLAM systems and datasets, offering broad compatibility and extendability while promoting reproducibility through consistent evaluation metrics and analysis tools. By reducing implementation complexity and minimizing configuration overhead, VSLAM-LAB empowers researchers to focus on advancing VSLAM methodologies and accelerates progress toward scalable, real-world solutions. We demonstrate the ease with which user-relevant benchmarks can be created: here, we introduce difficulty-level-based categories, but one could envision environment-specific or condition-specific categories.

Effective monitoring of underwater ecosystems is crucial for tracking environmental changes, guiding conservation efforts, and ensuring long-term ecosystem health. However, automating underwater ecosystem management with robotic platforms remains challenging due to the complexities of underwater imagery, which pose significant difficulties for traditional visual localization methods. We propose an integrated pipeline that combines Visual Place Recognition (VPR), feature matching, and image segmentation on images extracted from video sequences. This method enables robust identification of revisited areas, estimation of rigid transformations, and downstream analysis of ecosystem changes. Furthermore, we introduce the SQUIDLE+ VPR BenchmarkÐthe first large-scale underwater VPR benchmark designed to leverage an extensive collection of unstructured data from multiple robotic platforms, spanning time intervals from days to years. The dataset encompasses diverse trajectories with varying overlap and diverse seafloor types captured under different environmental conditions, including differences in depth, lighting, and turbidity. Our code is available at: https://github.com/bev-gorry/underloc.

In visual place recognition (VPR), filtering and sequence-based matching approaches can improve performance by integrating temporal information across image sequences, especially in challenging conditions. While these methods are commonly applied, their effects on system behavior can be unpredictable and can actually make performance worse in certain situations. In this work, we present a new supervised learning approach that learns to predict the per-frame sequence matching receptiveness (SMR) of VPR techniques, enabling the system to selectively decide when to trust the output of a sequence matching system. Our approach is agnostic to the underlying VPR technique and effectively predicts SMR, and hence significantly improves VPR performance across a large range of state-of-the-art and classical VPR techniques (namely CosPlace, MixVPR, EigenPlaces, SALAD, AP-GeM, NetVLAD and SAD), and across three benchmark VPR datasets (Nordland, Oxford RobotCar, and SFU-Mountain). We also provide insights into a complementary approach that uses the predictor to replace discarded matches, and present ablation studies including an analysis of the interactions between our SMR predictor and the selected sequence length.

Stand-alone Visual Place Recognition (VPR) systems have little defence against a well-designed adversarial attack, which can lead to disastrous consequences when deployed for robot navigation. This paper extensively analyzes the effect of two adversarial attacks common in other perception tasks and two novel VPR-specific attacks on VPR localization performance. We then propose how to close the loop between VPR, an Adversarial Attack Detector (AAD), and active navigation decisions by demonstrating the performance benefit of simulated AADs in a novel experiment paradigm – which we detail for the robotics community to use as a system framework. In the proposed experiment paradigm, we see the addition of AADs across a range of detection accuracies can improve performance over baseline; demonstrating a significant improvement – such as a ≈ 50% reduction in the mean along-track localization error – can be achieved with True Positive and False Positive detection rates of only 75% and up to 25% respectively. We examine a variety of metrics including: Along-Track Error, Percentage of Time Attacked, Percentage of Time in an ‘Unsafe’ State, and Longest Continuous Time Under Attack. Expanding further on these results, we provide the first investigation into the efficacy of the Fast Gradient Sign Method (FGSM) adversarial attack for VPR. The analysis in this work highlights the need for AADs in real-world systems for trustworthy navigation, and informs quantitative requirements for system design.

Domain adaptation is required for automated driving models to generalize well across diverse road conditions. This paper explores a training method for domain adaptation to adapt PilotNet, an end-to-end deep learning-based model, for left-hand driving conditions using real-world Australian highway data. Four training methods were evaluated: (1) a baseline model trained on U.S. right-hand driving data, (2) a model trained on flipped U.S. data, (3) a model pretrained on U.S. data and then fine-tuned on Australian highways, and (4) a model pretrained on flipped U.S. data and then fine-tuned on Australian highways. This setup examines whether incorporating flipped data enhances the model adaptation by providing an initial left-hand driving alignment. The paper compares model performance regarding steering prediction accuracy and attention, using saliency-based analysis to measure attention shifts across significant road regions. Results show that pretraining on flipped data alone worsens prediction stability due to misaligned feature representations, but significantly improves adaptation when followed by fine-tuning, leading to lower prediction error and stronger focus on left-side cues. To validate this approach across different architectures, the same experiments were done on ResNet, which confirmed similar adaptation trends. These findings emphasize the importance of preprocessing techniques, such as flipped-data pretraining, followed by fine-tuning to improve model adaptation with minimal retraining requirements.

In robotics, spiking neural networks (SNNs) are increasingly recognized for their largely unrealized potential energy efficiency and low latency particularly when implemented on neuromorphic hardware. This article highlights three advancements for SNNs in visual place recognition (VPR). First, we propose modular SNNs (Modular SNN), where each SNN represents a set of nonoverlapping geographically distinct places, enabling scalable networks for large environments. Second, we present ensembles of Modular SNNs, where multiple networks represent the same place, significantly enhancing accuracy compared to single-network models. Each of our Modular SNN modules is compact, comprising only 1500 neurons and 474k synapses, making them ideally suited for ensembling due to their small size. Finally, we investigate the role of sequence matching in SNN-based VPR, a technique where consecutive images are used to refine place recognition. We demonstrate competitive performance of our method on a range of datasets, including higher responsiveness to ensembling compared to conventional VPR techniques and higher R@1 improvements with sequence matching than VPR techniques with comparable baseline performance. Our contributions highlight the viability of SNNs for VPR, offering scalable and robust solutions, and paving the way for their application in various energy-sensitive robotic tasks.

State-of-the-art V-SLAM pipelines utilize robust cost functions and outlier rejection techniques to remove incorrect correspondences. However, these methods are typically fine-tuned to overfit certain benchmarks and struggle to adapt effectively to changes in the application domain or environmental conditions. This renders them impractical for many robotic applications in which robustness in a wide variety of conditions is essential. In this paper we introduce a novel distribution-based approach for online outlier rejection that reduces the necessity for scene-specific fine-tuning while simultaneously improving the overall SLAM performance. Through experiments across 3 different public datasets, we show that our approach consistently outperforms state-of-the-art methods in various real-world settings. Our code is available at https://github.com/alejandrofontan/ORB_SLAM2_Distribution

Spiking Neural Networks (SNNs) are at the forefront of neuromorphic computing thanks to their potential energy-efficiency, low latencies, and capacity for continual learning. While these capabilities are well suited for robotics tasks, SNNs have seen limited adaptation in this field thus far. This work introduces a SNN for Visual Place Recognition (VPR) that is both trainable within minutes and queryable in milliseconds, making it well suited for deployment on compute-constrained robotic systems. Our proposed system, VPRTempo, overcomes slow training and inference times using an abstracted SNN that trades biological realism for efficiency. VPRTempo employs a temporal code that determines the timing of a single spike based on a pixel’s intensity, as opposed to prior SNNs relying on rate coding that determined the number of spikes; improving spike efficiency by over 100%. VPRTempo is trained using Spike-Timing Dependent Plasticity and a supervised delta learning rule enforcing that each output spiking neuron responds to just a single place. We evaluate our system on the Nordland and Oxford RobotCar benchmark localization datasets, which include up to 27k places. We found that VPRTempo’s accuracy is comparable to prior SNNs and the popular NetVLAD place recognition algorithm, while being several orders of magnitude faster and suitable for real-time deployment – with inference speeds over 50 Hz on CPU. VPRTempo could be integrated as a loop closure component for online SLAM on resource-constrained systems such as space and underwater robots.

Robot navigation requires an autonomy pipeline that is robust to environmental changes and effective in varying conditions. Teach and Repeat (T&R) navigation has shown high performance in autonomous repeated tasks under challenging circumstances, but research within T&R has predominantly focused on motion planning as opposed to motion control. In this paper, we propose a novel T&R system based on a robust motion control technique for a skid-steering mobile robot using sliding-mode control that effectively handles uncertainties that are particularly pronounced in the T&R task, where sensor noises, parametric uncertainties, and wheel-terrain interaction are common challenges. We first theoretically demonstrate that the proposed T&R system is globally stable and robust while considering the uncertainties of the closed-loop system. When deployed on a Clearpath Jackal robot, we then show the global stability of the proposed system in both indoor and outdoor environments covering different terrains, outperforming previous state-of-the-art methods in terms of mean average trajectory error and stability in these challenging environments. This paper makes an important step towards long-term autonomous T&R navigation with ensured safety guarantees.

Mobile robots and autonomous vehicles are often required to function in environments where critical position estimates from sensors such as GPS become uncertain or unreliable. Single image visual place recognition (VPR) provides an alternative for localization but often requires techniques such as sequence matching to improve robustness, which incurs additional computation and latency costs. Even then, the sequence length required to localize at an acceptable performance level varies widely; and simply setting overly long fixed sequence lengths creates unnecessary latency, computational overhead, and can even degrade performance. In these scenarios it is often more desirable to meet or exceed a set target performance at minimal expense. In this paper we present an approach which uses a calibration set of data to fit a model that modulates sequence length for VPR as needed to exceed a target localization performance. We make use of a coarse position prior, which could be provided by any other localization system, and capture the variation in appearance across this region. We use the correlation between appearance variation and sequence length to curate VPR features and fit a Multi-Layer Perceptron (MLP) for selecting the optimal length. We demonstrate that this method is effective at modulating sequence length to maximize the number of sections in a dataset which meet or exceed a target performance whilst minimizing the median length used. We show applicability across several datasets and reveal key phenomena like generalization capabilities, the benefits of curating features and the utility of non-state-of-the-art feature extractors with nuanced properties.

Event cameras are increasingly popular in robotics due to beneficial features such as low latency, energy efficiency, and high dynamic range. Nevertheless, their downstream task performance is greatly influenced by the optimization of bias parameters. These parameters, for instance, regulate the necessary change in light intensity to trigger an event, which in turn depends on factors such as the environment lighting and camera motion. This paper introduces feedback control algorithms that automatically tune the bias parameters through two interacting methods: 1) An immediate, on-the-fly fast adaptation of the refractory period, which sets the minimum interval between consecutive events, and 2) if the event rate exceeds the specified bounds even after changing the refractory period repeatedly, the controller adapts the pixel bandwidth and event thresholds, which stabilizes after a short period of noise events across all pixels (slow adaptation). Our evaluation focuses on the visual place recognition task, where incoming query images are compared to a given reference database. We conducted comprehensive evaluations of our algorithms’ adaptive feedback control in real-time. To do so, we collected the QCR-Fast-and-Slow dataset that contains DAVIS346 event camera streams from 366 repeated traversals of a Scout Mini robot navigating through a 100 meter long indoor lab setting (totaling over 35km distance traveled) in varying brightness conditions with ground truth location information. Our proposed feedback controllers result in superior performance when compared to the standard bias settings and prior feedback control methods. Our findings also detail the impact of bias adjustments on task performance and feature ablation studies on the fast and slow adaptation mechanisms.

We propose FocusTune, a focus-guided sampling technique to improve the performance of visual localization algorithms. FocusTune directs a scene coordinate regression model towards regions critical for 3D point triangulation by exploiting key geometric constraints. Specifically, rather than uniformly sampling points across the image for training the scene coordinate regression model, we instead re-project 3D scene coordinates onto the 2D image plane and sample within a local neighborhood of the reprojected points. While our proposed sampling strategy is generally applicable, we showcase FocusTune by integrating it with the recently introduced Accelerated Coordinate Encoding (ACE) model. Our results demonstrate that FocusTune both improves or matches state-of-the-art performance whilst keeping ACE’s appealing low storage and compute requirements, for example reducing translation error from 25 to 19 and 17 to 15 cm for single and ensemble models, respectively, on the Cambridge Landmarks dataset. This combination of high performance and low compute and storage requirements is particularly promising for applications in areas like mobile robotics and augmented reality. We made our code available at https://github.com/sontung/focus-tune.

While substantial progress has been made in the absolute performance of localization and Visual Place Recognition (VPR) techniques, it is becoming increasingly clear from translating these systems into applications that other capabilities like integrity and predictability are just as important, especially for safety- or operationally-critical autonomous systems. In this research we present a new, training-free approach to predicting the likely quality of localization estimates, and a novel method for using these predictions to bias a sequence-matching process to produce additional performance gains beyond that of a naive sequence matching approach. Our combined system is lightweight, runs in real-time and is agnostic to the underlying VPR technique. On extensive experiments across four datasets and three VPR techniques, we demonstrate our system improves precision performance, especially at the high-precision/low-recall operating point. We also present ablation and analysis identifying the performance contributions of the prediction and weighted sequence matching components in isolation, and the relationship between the quality of the prediction system and the benefits of the weighted sequential matcher.

Spiking neural networks have significant potential utility in robotics due to their high energy efficiency on specialized hardware, but proof-of-concept implementations have not yet typically achieved competitive performance or capability with conventional approaches. In this paper, we tackle one of the key practical challenges of scalability by introducing a novel modular ensemble network approach, where compact, localized spiking networks each learn and are solely responsible for recognizing places in a local region of the environment only. This modular approach creates a highly scalable system. However, it comes with a high-performance cost where a lack of global regularization at deployment time leads to hyperactive neurons that erroneously respond to places outside their learned region. Our second contribution introduces a regularization approach that detects and removes these problematic hyperactive neurons during the initial environmental learning phase. We evaluate this new scalable modular system on benchmark localization datasets Nordland and Oxford RobotCar, with comparisons to standard techniques NetVLAD, DenseVLAD, and SAD, and a previous spiking neural network system. Our system substantially outperforms the previous SNN system on its small dataset, but also maintains performance on 27 times larger benchmark datasets where the operation of the previous system is computationally infeasible, and performs competitively with the conventional localization systems.

One recent promising approach to the Visual Place Recognition (VPR) problem has been to fuse the place recognition estimates of multiple complementary VPR techniques using methods such as shared representative appearance learning (SRAL) and multi-process fusion. These approaches come with a substantial practical limitation: they require all potential VPR methods to be brute-force run before they are selectively fused. The obvious solution to this limitation is to predict the viable subset of methods ahead of time, but this is challenging because it requires a predictive signal within the imagery itself that is indicative of high performance methods. Here we propose an alternative approach that instead starts with a known single base VPR technique, and learns to predict the most complementary additional VPR technique to fuse with it, that results in the largest improvement in performance. The key innovation here is to use a dimensionally reduced difference vector between the query image and the top-retrieved reference image using this baseline technique as the predictive signal of the most complementary additional technique, both during training and inference. We demonstrate that our approach can train a single network to select performant, complementary technique pairs across datasets which span multiple modes of transportation (train, car, walking) as well as to generalise to unseen datasets, outperforming multiple baseline strategies for manually selecting the best technique pairs based on the same training data.

Surgical assistive robots offer the potential for drastically improved patient outcomes through more accurate, more repeatable surgical procedures like shoulder arthroplasty operations. Existing robotic systems typically rely on optical marker tracking and require invasive marker attachment for localization, complicating the surgical workflow and patient recovery. But moving towards a markerless system is very challenging, both because of the absolute difficulty and the large variation in localization conditions across thousands of surgical procedures. In this paper we propose an alternative approach: rather than try to create a “perfect” and fully generalizable markerless localization system, instead create a reliable and trustworthy localization system that is able to continually self-assess the likely quality of its localization esti-mates, and act accordingly. We propose a lightweight method for predicting vision-based pose estimation performance using internal pipeline artifacts (without needing external ground truth from a marker-based system). Using extensive real robot experiments with challenging actual imagery from surgery, we demonstrate our prediction system accurately self-characterizes the localization system's performance across a wide range of localization conditions, and demonstrate that this prediction system generalizes to a range of surgical conditions. We then show how online performance prediction can drive active robot navigation that minimizes localization error, reducing target pose estimation error by 96.1% for rotation and 96.7% for translation compared to rejected alternative trajectories.

Can knowing where you are assist in perceiving objects in your surroundings, especially under adverse weather and lighting conditions? In this work we investigate whether a prior map can be leveraged to aid in the detection of dynamic objects in a scene without the need for a 3D map or pixel-level map-query correspondences. We contribute an algorithm which refines an initial set of candidate object detections and produces a refined subset of highly accurate detections using a prior map. We begin by using visual place recognition (VPR) to retrieve a prior map image for a given query image, then use a binary classification neural network that compares the query and prior map image regions to validate the query detection. Once our classification network is trained, on approximately 1000 query-map image pairs, it is able to improve the performance of vehicle detection when combined with an existing off-the-shelf vehicle detector. We demonstrate our approach using standard datasets across two cities (Oxford and Zurich) under different settings of train-test separation of map-query traverse pairs. We further emphasize the performance gains of our approach against alternative design choices and show that VPR suffices for the task, eliminating the need for precise ground truth localization.

Most 6-DoF localization and SLAM systems use static landmarks but ignore dynamic objects because they cannot be usefully incorporated into a typical pipeline. Where dynamic objects have been incorporated, typical approaches have attempted relatively sophisticated identification and localization of these objects, limiting their robustness or general utility. In this research, we propose a middle ground, demonstrated in the context of autonomous vehicles, using dynamic vehicles to provide limited pose constraint information in a 6-DoF frame-by-frame PnP-RANSAC localization pipeline. We refine initial pose estimates with a motion model and propose a method for calculating the predicted quality of future pose estimates, triggered by whether or not the autonomous vehicle's motion is constrained by the relative frame-to-frame location of dynamic vehicles in the environment. Our approach detects and identifies suitable dynamic vehicles to define these pose constraints to modify a pose filter, resulting in improved recall across a range of localization tolerances from 0.25m to 5m, compared to a state-of-the-art baseline single image PnP method and its vanilla pose filtering. Our constraint detection system is active for approximately 35% of the time on the Ford AV dataset and localization is particularly improved when the constraint detection is active.

Animals and insects showcase remarkably robust and adept navigational abilities, up to literally circumnavigating the globe. Primary progress in robotics inspired by these natural systems has occurred in two areas: highly theoretical computational neuroscience models, and handcrafted systems like RatSLAM and NeuroSLAM. In this research, we present work bridging the gap between the two, in the form of Multiscale Continuous Attractor Networks (MCAN), that combine the multiscale parallel spatial neural networks of the previous theoretical models with the real-world robustness of the robot-targeted systems, to enable trajectory tracking over large velocity ranges. To overcome the limitations of the reliance of previous systems on hand-tuned parameters, we present a genetic algorithm-based approach for automated tuning of these networks, substantially improving their usability. To provide challenging navigational scale ranges, we open source a flexible city-scale navigation simulator that adapts to any street network, enabling high throughput experimentation11https://github.com/theresejoseph/Trajectory_ Tracking_via_MCAN/ • In extensive experiments using the city-scale navigation environment and Kitti, we show that the system is capable of stable dead reckoning over a wide range of velocities and environmental scales, where a single-scale approach fails.

Recently several fusion and switching based approaches have been presented to solve the problem of Visual Place Recognition. In spite of these systems demonstrating significant boost in VPR performance they each have their own set of limitations. The multi-process fusion systems usually involve employing brute force and running all available VPR techniques simultaneously while the switching method attempts to negate this practise by only selecting the best suited VPR technique for given query image. But switching does fail at times when no available suitable technique can be identified. An innovative solution would be an amalgamation of the two otherwise discrete approaches to combine their competitive advantages while negating their shortcomings. The proposed, Switch-Fuse system, is an interesting way to combine both the robustness of switching VPR techniques based on complementarity and the force of fusing the carefully selected techniques to significantly improve performance. Our system holds a structure superior to the basic fusion methods as instead of simply fusing all or any random techniques, it is structured to first select the best possible VPR techniques for fusion, according to the query image. The system combines two significant processes, switching and fusing VPR techniques, which together as a hybrid model substantially improve performance on all major VPR data sets illustrated using PR curves.

Scene classification is a well-established area of computer vision research that aims to classify a scene image into pre-defined categories such as playground, beach and airport. Recent work has focused on increasing the variety of pre-defined categories for classification, but so far failed to consider two major challenges: changes in scene appearance due to lighting and open set classification (the ability to classify unknown scene data as not belonging to the trained classes). Our first contribution, SceneVLAD, fuses scene classification and visual place recognition CNNs for appearance invariant scene classification that outperforms state-of-the-art scene classification by a mean F1 score of up to 0.1. Our second contribution, OpenSceneVLAD, extends the first to an open set classification scenario using intra-class splitting to achieve a mean increase in F1 scores of up to 0.06 compared to using state-of-the-art openmax layer. We achieve these results on three scene class datasets extracted from large scale outdoor visual localisation datasets, one of which we collected ourselves.

VPR is a fundamental task for autonomous navigation as it enables a robot to localize itself in the workspace when a known location is detected. Although accuracy is an essential requirement for a VPR technique, computational and energy efficiency are not less important for real-world applications. CNN-based techniques archive state-of-the-art VPR performance but are computationally intensive and energy demanding. Binary neural networks (BNN) have been recently proposed to address VPR efficiently. Although a typical BNN is an order of magnitude more efficient than a CNN, its processing time and energy usage can be further improved. In a typical BNN, the first convolution is not completely binarized for the sake of accuracy. Consequently, the first layer is the slowest network stage, requiring a large share of the entire computational effort. This paper presents a class of BNNs for VPR that combines depthwise separable factorization and binarization to replace the first convolutional layer to improve computational and energy efficiency. Our best model achieves higher VPR performance while spending considerably less time and energy to process an image than a BNN using a non-binary convolution as a first stage.

Visual place recognition (VPR) - a fundamental task in computer vision and robotics - is the problem of identifying a place mainly based on visual information. View-point and appearance changes, such as due to weather and seasonal variations, make this task challenging. Currently, there is no universal VPR technique that can work in all types of environments, on a variety of robotic platforms, and under a wide range of viewpoint and appearance changes. Recent work has shown the potential of combining different VPR methods intelligently by evaluating complementarity for some specific VPR datasets to achieve better performance. This, however, requires ground truth information (correct matches) which is not available when a robot is deployed in a real-world scenario. Moreover, running multiple VPR techniques in parallel may be prohibitive for resource-constrained embedded platforms. To overcome these limitations, this paper presents a probabilistic complementarity-based switching VPR system, SwitchHit. Our proposed system consists of multiple VPR techniques, however, it does not simply run all techniques at once, rather predicts the probability of correct match for an incoming query image and dynamically switches to another complementary technique if the probability of correctly matching the query is below a certain threshold. This innovative use of multiple VPR techniques allow our system to be more efficient and robust than other combined VPR approaches employing brute force and running multiple VPR techniques at once. Thus making it more suitable for resource constrained embedded systems and achieving an overall superior performance from what any individual VPR method in the system could have by achieved running independently.

Visual place recognition (VPR) is a robot’s ability to determine whether a place was visited before using visual data. While conventional handcrafted methods for VPR fail under extreme environmental appearance changes, those based on convolutional neural networks (CNNs) achieve state-of-the-art performance but result in heavy runtime processes and model sizes that demand a large amount of memory. Hence, CNN-based approaches are unsuitable for resource-constrained platforms, such as small robots and drones. In this article, we take a multistep approach of decreasing the precision of model parameters, combining it with network depth reduction and fewer neurons in the classifier stage to propose a new class of highly compact models that drastically reduces the memory requirements and computational effort while maintaining state-of-the-art VPR performance. To the best of our knowledge, this is the first attempt to propose binary neural networks for solving the VPR problem effectively under changing conditions and with significantly reduced resource requirements. Our best-performing binary neural network, dubbed FloppyNet, achieves comparable VPR performance when considered against its full-precision and deeper counterparts while consuming 99% less memory and increasing the inference speed by seven times.

Visual Place Recognition is a challenging task for robotics and autonomous systems, which must deal with the twin problems of appearance and viewpoint change in an always changing world. This paper introduces Patch-NetVLAD, which provides a novel formulation for combining the advantages of both local and global descriptor methods by deriving patch-level features from NetVLAD residuals. Unlike the fixed spatial neighborhood regime of existing local keypoint features, our method enables aggregation and matching of deep-learned local features defined over the feature-space grid. We further introduce a multi-scale fusion of patch features that have complementary scales (i.e. patch sizes) via an integral feature space and show that the fused features are highly invariant to both condition (season, structure, and illumination) and viewpoint (translation and rotation) changes. Patch-NetVLAD achieves state-of-the-art visual place recognition results in computationally limited scenarios, validated on a range of challenging real-world datasets, including winning the Facebook Mapillary Visual Place Recognition Challenge at ECCV2020. It is also adaptable to user requirements, with a speed-optimised version operating over an order of magnitude faster than the state-of-the-art. By combining superior performance with improved computational efficiency in a configurable framework, Patch-NetVLAD is well suited to enhance both stand-alone place recognition capabilities and the overall performance of SLAM systems.

We explore the zero-shot setting for day-night domain adaptation. The traditional domain adaptation setting is to train on one domain and adapt to the target domain by exploiting unlabeled data samples from the test set. As gathering relevant test data is expensive and sometimes even impossible, we remove any reliance on test data imagery and instead exploit a visual inductive prior derived from physics-based reflection models for domain adaptation. We cast a number of color invariant edge detectors as trainable layers in a convolutional neural network and evaluate their robustness to illumination changes. We show that the color invariant layer reduces the day-night distribution shift in feature map activations throughout the network. We demonstrate improved performance for zero-shot day to night domain adaptation on both synthetic as well as natural datasets in various tasks, including classification, segmentation and place recognition.

Visual Place Recognition (VPR) is often characterized as being able to recognize the same place despite significant changes in appearance and viewpoint. VPR is a key component of Spatial Artificial Intelligence, enabling robotic platforms and intelligent augmentation platforms such as augmented reality devices to perceive and understand the physical world. In this paper, we observe that there are three "drivers" that impose requirements on spatially intelligent agents and thus VPR systems: 1) the particular agent including its sensors and computational resources, 2) the operating environment of this agent, and 3) the specific task that the artificial agent carries out. In this paper, we characterize and survey key works in the VPR area considering those drivers, including their place representation and place matching choices. We also provide a new definition of VPR based on the visual overlap - akin to spatial view cells in the brain - that enables us to find similarities and differences to other research areas in the robotics and computer vision fields. We identify several open challenges and suggest areas that require more in-depth attention in future works.

Fully autonomous mobile robots have a multitude of potential applications, but guaranteeing robust navigation performance remains an open research problem. For many tasks such as repeated infrastructure inspection, item delivery, or inventory transport, a route repeating capability can be sufficient and offers potential practical advantages over a full navigation stack. Previous teach and repeat research has achieved high performance in difficult conditions predominantly by using sophisticated, expensive sensors, and has often had high computational requirements. Biological systems, such as small animals and insects like seeing ants, offer a proof of concept that robust and generalisable navigation can be achieved with extremely limited visual systems and computing power. In this work we create a novel asynchronous formulation for teach and repeat navigation that fully utilises odometry information, paired with a correction signal driven by much more computationally lightweight visual processing than is typically required. This correction signal is also decoupled from the robot’s motor control, allowing its rate to be modulated by the available computing capacity. We evaluate this approach with extensive experimentation on two different robotic platforms, the Consequential Robotics Miro and the Clearpath Jackal robots, across navigation trials totalling more than 6000 metres in a range of challenging indoor and outdoor environments. Our approach continues to succeed when multiple state-of-the-art systems fail due to low resolution images, unreliable odometry, or lighting change, while requiring significantly less compute. We also – for the first time – demonstrate versatile cross-platform teach and repeat without changing parameters, in which we learn to navigate a route with one robot and repeat that route using a completely different robot.

The use of local detectors and descriptors in typical computer vision pipelines works well until variations in viewpoint and appearance change become extreme. Past research in this area has typically focused on one of two approaches to this challenge: the use of projections into spaces more suitable for feature matching under extreme viewpoint changes, and attempting to learn features that are inherently more robust to viewpoint change. In this paper, we present a novel framework that combines the learning of invariant descriptors through data augmentation and orthographic viewpoint projection. We propose rotation-robust local descriptors, learnt through training data augmentation based on rotation homographies, and a correspondence ensemble technique that combines vanilla feature correspondences with those obtained through rotation-robust features. Using a range of benchmark datasets as well as contributing a new bespoke dataset for this research domain, we evaluate the effectiveness of the proposed approach on key tasks including pose estimation and visual place recognition. Our system outperforms a range of baseline and state-of-the-art techniques, including enabling higher levels of place recognition precision across opposing place viewpoints, and achieves practically useful performance levels even under extreme viewpoint changes. We reduce pose estimation error by 86.72% relative to state of the art.

In open set recognition, deep neural networks encounter object classes that were unknown during training. Existing open set classifiers distinguish between known and unknown classes by measuring distance in a network's logit space, assuming that known classes cluster closer to the training data than unknown classes. However, this approach is applied post-hoc to networks trained with cross-entropy loss, which does not guarantee this clustering behaviour. To overcome this limitation, we introduce the Class Anchor Clustering (CAC) loss. CAC is a distance-based loss that explicitly trains known classes to form tight clusters around anchored class-dependent centres in the logit space. We show that training with CAC achieves state-of-the-art performance for distance-based open set classifiers on all six standard benchmark datasets, with a 15.2% AUROC increase on the challenging TinyImageNet, without sacrificing classification accuracy. We also show that our anchored class centres achieve higher open set performance than learnt class centres, particularly on object-based datasets and large numbers of training classes.

Abstract Visual place recognition (VPR) is the process of recognising a previously visited place using visual information, often under varying appearance conditions and viewpoint changes and with computational constraints. VPR is related to the concepts of localisation, loop closure, image retrieval and is a critical component of many autonomous navigation systems ranging from autonomous vehicles to drones and computer vision systems. While the concept of place recognition has been around for many years, VPR research has grown rapidly as a field over the past decade due to improving camera hardware and its potential for deep learning-based techniques, and has become a widely studied topic in both the computer vision and robotics communities. This growth however has led to fragmentation and a lack of standardisation in the field, especially concerning performance evaluation. Moreover, the notion of viewpoint and illumination invariance of VPR techniques has largely been assessed qualitatively and hence ambiguously in the past. In this paper, we address these gaps through a new comprehensive open-source framework for assessing the performance of VPR techniques, dubbed “VPR-Bench”. VPR-Bench (Open-sourced at: https://github.com/MubarizZaffar/VPR-Bench ) introduces two much-needed capabilities for VPR researchers: firstly, it contains a benchmark of 12 fully-integrated datasets and 10 VPR techniques, and secondly, it integrates a comprehensive variation-quantified dataset for quantifying viewpoint and illumination invariance. We apply and analyse popular evaluation metrics for VPR from both the computer vision and robotics communities, and discuss how these different metrics complement and/or replace each other, depending upon the underlying applications and system requirements. Our analysis reveals that no universal SOTA VPR technique exists, since: (a) state-of-the-art (SOTA) performance is achieved by 8 out of the 10 techniques on at least one dataset, (b) SOTA technique in one community does not necessarily yield SOTA performance in the other given the differences in datasets and metrics. Furthermore, we identify key open challenges since: (c) all 10 techniques suffer greatly in perceptually-aliased and less-structured environments, (d) all techniques suffer from viewpoint variance where lateral change has less effect than 3D change, and (e) directional illumination change has more adverse effects on matching confidence than uniform illumination change. We also present detailed meta-analyses regarding the roles of varying ground-truths, platforms, application requirements and technique parameters. Finally, VPR-Bench provides a unified implementation to deploy these VPR techniques, metrics and datasets, and is extensible through templates.

This paper presents a cognition-inspired agnostic framework for building a map for Visual Place Recognition. This framework draws inspiration from human-memorability, utilizes the traditional image entropy concept and computes the static content in an image; thereby presenting a tri-folded criteria to assess the ‘memorability’ of an image for visual place recognition. A dataset namely ‘ESSEX3IN1’ is created, composed of highly confusing images from indoor, outdoor and natural scenes for analysis. When used in conjunction with state-of-the-art visual place recognition methods, the proposed framework provides significant performance boost to these techniques, as evidenced by results on ESSEX3IN1 and other public datasets.

Visual navigation tasks in real-world environments often require both self-motion and place recognition feedback. While deep reinforcement learning has shown success in solving these perception and decision-making problems in an end-to-end manner, these algorithms require large amounts of experience to learn navigation policies from high-dimensional data, which is generally impractical for real robots due to sample complexity. In this paper, we address these problems with two main contributions. We first leverage place recognition and deep learning techniques combined with goal destination feedback to generate compact, bimodal image representations that can then be used to effectively learn control policies from a small amount of experience. Second, we present an interactive framework, CityLearn, that enables for the first time training and deployment of navigation algorithms across city-sized, realistic environments with extreme visual appearance changes. CityLearn features more than 10 benchmark datasets, often used in visual place recognition and autonomous driving research, including over 100 recorded traversals across 60 cities around the world. We evaluate our approach on two CityLearn environments, training our navigation policy on a single traversal per dataset. Results show our method can be over 2 orders of magnitude faster than when using raw images, and can also generalize across extreme visual changes including day to night and summer to winter transitions.

Visual place recognition algorithms trade off three key characteristics: their storage footprint, their computational requirements, and their resultant performance, often expressed in terms of recall rate. Significant prior work has investigated highly compact place representations, sub-linear computational scaling and sub-linear storage scaling techniques, but have always involved a significant compromise in one or more of these regards, and have only been demonstrated on relatively small datasets. In this paper we present a novel place recognition system which enables for the first time the combination of ultra-compact place representations, near sub-linear storage scaling and extremely lightweight compute requirements. Our approach exploits the inherently sequential nature of much spatial data in the robotics domain and inverts the typical target criteria, through intentionally coarse scalar quantization-based hashing that leads to more collisions but is resolved by sequence-based matching. For the first time, we show how effective place recognition rates can be achieved on a new very large 10 million place dataset, requiring only 8 bytes of storage per place and 37K unitary operations to achieve over 50% recall for matching a sequence of 100 frames, where a conventional stateof-the-art approach both consumes 1300 times more compute and fails catastrophically. We present analysis investigating the effectiveness of our hashing overload approach under varying sizes of quantized vector length, comparison of near miss matches with the actual match selections and characterise the effect of variance re-scaling of data on quantization. Resource link: https://github.com/oravus/CoarseHash.

Combining multiple complementary techniques together has long been regarded as a way to improve performance. In visual localization, multi-sensor fusion, multi-process fusion of a single sensing modality, and even combinations of different localization techniques have been shown to result in improved performance. However, merely fusing together different localization techniques does not account for the varying performance characteristics of different localization techniques. In this paper we present a novel, hierarchical localization system that explicitly benefits from three varying characteristics of localization techniques: the distribution of their localization hypotheses, their appearance- and viewpointinvariant properties, and the resulting differences in where in an environment each system works well and fails. We show how two techniques deployed hierarchically work better than in parallel fusion, how combining two different techniques works better than two levels of a single technique, even when the single technique has superior individual performance, and develop two and three-tier hierarchical structures that progressively improve localization performance. Finally, we develop a stacked hierarchical framework where localization hypotheses from techniques with complementary characteristics are concatenated at each layer, significantly improving retention of the correct hypothesis through to the final localization stage. Using two challenging datasets, we show the proposed system outperforming state-of-the-art techniques.

In this work we focus on improving the efficiency and generalisation of learned navigation strategies when transferred from its training environment to previously unseen ones. We present an extension of the residual reinforcement learning framework from the robotic manipulation literature and adapt it to the vast and unstructured environments that mobile robots can operate in. The concept is based on learning a residual control effect to add to a typical sub-optimal classical controller in order to close the performance gap, whilst guiding the exploration process during training for improved data efficiency. We exploit this tight coupling and propose a novel deployment strategy, switching Residual Reactive Navigation (sRRN), which yields efficient trajectories whilst probabilistically switching to a classical controller in cases of high policy uncertainty. Our approach achieves improved performance over end-to-end alternatives and can be incorporated as part of a complete navigation stack for cluttered indoor navigation tasks in the real world. The code and training environment for this project is made publicly available at https://sites.google.com/view/srrn/home.

Learning-based approaches often outperform hand-coded algorithmic solutions for many problems in robotics. However, learning long-horizon tasks on real robot hardware can be intractable, and transferring a learned policy from simulation to reality is still extremely challenging. We present a novel approach to model-free reinforcement learning that can leverage existing sub-optimal solutions as an algorithmic prior during training and deployment. During training, our gated fusion approach enables the prior to guide the initial stages of exploration, increasing sample-efficiency and enabling learning from sparse long-horizon reward signals. Importantly, the policy can learn to improve beyond the performance of the sub-optimal prior since the prior's influence is annealed gradually. During deployment, the policy's uncertainty provides a reliable strategy for transferring a simulation-trained policy to the real world by falling back to the prior controller in uncertain states. We show the efficacy of our Multiplicative Controller Fusion approach on the task of robot navigation and demonstrate safe transfer from simulation to the real world without any fine-tuning. The code for this project is made publicly available at https://sites.google.com/view/mcf-nav/home.

This article presents a lightweight visual place recognition approach, capable of achieving high performance with low computational cost, and feasible for mobile robotics under significant viewpoint and appearance changes. Results on several benchmark datasets confirm an average boost of 13% in accuracy, and 12x average speedup relative to state-of-the-art methods.

Visual place recognition (VPR) - the act of recognizing a familiar visual place - becomes difficult when there is extreme environmental appearance change or viewpoint change. Particularly challenging is the scenario where both phenomena occur simultaneously, such as when returning for the first time along a road at night that was previously traversed during the day in the opposite direction. While such problems can be solved with panoramic sensors, humans solve this problem regularly with limited field-of-view vision and without needing to constantly turn around. In this paper, we present a new depth- and temporal-aware visual place recognition system that solves the opposing viewpoint, extreme appearance-change visual place recognition problem. Our system performs sequence-to-single frame matching by extracting depth-filtered keypoints using a state-of-the-art depth estimation pipeline, constructing a keypoint sequence over multiple frames from the reference dataset, and comparing these keypoints to the keypoints extracted from a single query image. We evaluate the system on a challenging benchmark dataset and show that it consistently outperforms state-of-the-art techniques. We also develop a range of diagnostic simulation experiments that characterize the contribution of depth-filtered keypoint sequences with respect to key domain parameters including the degree of appearance change and camera motion.

There has been a recent emergence of sampling-based techniques for estimating epistemic uncertainty in deep neural networks. While these methods can be applied to classification or semantic segmentation tasks by simply averaging samples, this is not the case for object detection, where detection sample bounding boxes must be accurately associated and merged. A weak merging strategy can significantly degrade the performance of the detector and yield an unreliable uncertainty measure. This paper provides the first in-depth investigation of the effect of different association and merging strategies. We compare different combinations of three spatial and two semantic affinity measures with four clustering methods for MC Dropout with a Single Shot Multi-Box Detector. Our results show that the correct choice of affinity-clustering combination can greatly improve the effectiveness of the classification and spatial uncertainty estimation and the resulting object detection performance. We base our evaluation on a new mix of datasets that emulate near open-set conditions (semantically similar unknown classes), distant open-set conditions (semantically dissimilar unknown classes) and the common closed-set conditions (only known classes).

A key capability for autonomous underground mining vehicles is real-time accurate localisation. While significant progress has been made, currently deployed systems have several limitations ranging from dependence on costly additional infrastructure to failure of both visual and range-sensor-based techniques in highly aliased or visually challenging environments. In our previous work, we presented a lightweight coarse vision-based localisation system that could map and then localise to within a few metres in an underground mining environment. However, this level of precision is insufficient for providing a cheaper, more reliable vision-based automation alternative to current range sensor-based systems. Here we present a new precision localisation system dubbed “LookUP”, which learns a neural-network-based pixel sampling strategy for estimating homographies based on ceiling-facing cameras without requiring any manual labelling. This new system runs in real time on limited computation resource and is demonstrated on two different underground mine sites, achieving real time performance at ~5 frames per second and a much improved average localisation error of ~1.2 metre.

CNNs have excelled at performing place recognition over time, particularly when the neural network is optimized for localization in the current environmental conditions. In this paper we investigate the concept of feature map filtering, where, rather than using all the activations within a convolutional tensor, only the most useful activations are used. Since specific feature maps encode different visual features, the objective is to remove feature maps that detract from the ability to recognize a location across appearance changes. Our key innovation is to filter the feature maps in an early convolutional layer, but then continue to run the network and extract a feature vector using a later, more viewpoint invariant layer in the same network. Our approach improves the condition and viewpoint invariance of a pre-trained network, using as little as a single training image pair from the deployment environment. An exhaustive experimental analysis is performed to determine the full scope of causality between early layer filtering and late layer extraction. For validity, we use three datasets: Oxford RobotCar, Nordland, and Gardens Point, achieving overall superior performance to NetVLAD. The work provides a number of new avenues for exploring CNN optimizations, without requiring any re-training of the network weights.

When a human drives a car along a road for the first time, they later recognize where they are on the return journey typically without needing to look in their rear view mirror or turn around to look back, despite significant viewpoint and appearance change. Such navigation capabilities are typically attributed to our semantic visual understanding of the environment [1] beyond geometry to recognizing the types of places we are passing through such as “passing a shop on the left” or “moving through a forested area”. Humans are in effect using place categorization [2] to perform specific place recognition even when the viewpoint is 180 degrees reversed. Recent advances in deep neural networks have enabled high performance semantic understanding of visual places and scenes, opening up the possibility of emulating what humans do. In this work, we develop a novel methodology for using the semantics-aware higher-order layers of deep neural networks for recognizing specific places from within a reference database. To further improve the robustness to appearance change, we develop a descriptor normalization scheme that builds on the success of normalization schemes for pure appearance-based techniques such as SeqSLAM [3]. Using two different datasets - one road-based, one pedestrian-based, we evaluate the performance of the system in performing place recognition on reverse traversals of a route with a limited field of view camera and no turn-back-and-Iook behaviours, and compare to existing state-of-the-art techniques and vanilla off-the-shelf features. The results demonstrate significant improvements over the existing state of the art, especially for extreme perceptual challenges that involve both great viewpoint change and environmental appearance change. We also provide experimental analyses of the contributions of the various system components: the use of spatio-temporal sequences, place categorization and place-centric characteristics as opposed to object-centric semantics.

Place recognition is an essential component of Simultaneous Localization And Mapping (SLAM). Under severe appearance change, reliable place recognition is a difficult perception task since the same place is perceptually very different in the morning, at night, or over different seasons. This work addresses place recognition as a domain translation task. Using a pair of coupled Generative Adversarial Networks (GANs), we show that it is possible to generate the appearance of one domain (such as summer) from another (such as winter) without requiring image-to-image correspondences across the domains. Mapping between domains is learned from sets of images in each domain without knowing the instance-to-instance correspondence by enforcing a cyclic consistency constraint. In the process, meaningful feature spaces are learned for each domain, the distances in which can be used for the task of place recognition. Experiments show that learned features correspond to visual similarity and can be effectively used for place recognition across seasons.

This work presents Semi-Supervised SLAM - a method for developing a map suitable for coarse localization within an underground environment with minimal human intervention, with system characteristics driven by real-world requirements of major mining companies. This work leverages existing information common within a mining environment - namely a surveyed mine map - which is used to sparsely ground map locations within the mine environment, increasing map accuracy and allowing localization within a global frame. Map creation utilizes a low cost camera sensor and minimal user information to produce a map which can be used for single camera localization within a mining environment. We evaluate the localization capabilities of the proposed approach in depth by performing data collection on operational underground mining vehicles within an active underground mine and by simulating occlusions common to the environment such as dust and water. The proposed system is capable of producing maps which have an average localization error 2.5 times smaller than the next best performing method ORB-SLAM2, comparable localization performance to a state-of-the-art deep learning approach (which is not a feasible solution due to both compute and training requirements) and is robust to simulated environmental obscurants.

Visually recognising a traversed route - regardless of whether seen during the day or night, in clear or inclement conditions, or in summer or winter - is an important capability for navigating robots. Since SeqSLAM was introduced in 2012, a large body of work has followed exploring how robotic systems can use the algorithm to meet the challenges posed by navigation in changing environmental conditions. The following paper describes OpenSeqSLAM2.0, a fully open-source toolbox for visual place recognition under changing conditions. Beyond the benefits of open access to the source code, OpenSeqSLAM2.0 provides a number of tools to facilitate exploration of the visual place recognition problem and interactive parameter tuning. Using the new open source platform, it is shown for the first time how comprehensive parameter characterisations provide new insights into many of the system components previously presented in ad hoc ways and provide users with a guide to what system component options should be used under what circumstances and why.

The success of deep learning techniques in the computer vision domain has triggered a range of initial investigations into their utility for visual place recognition, all using generic features from networks that were trained for other types of recognition tasks. In this paper, we train, at large scale, two CNN architectures for the specific place recognition task and employ a multi-scale feature encoding method to generate condition- and viewpoint-invariant features. To enable this training to occur, we have developed a massive Specific PlacEs Dataset (SPED) with hundreds of examples of place appearance change at thousands of different places, as opposed to the semantic place type datasets currently available. This new dataset enables us to set up a training regime that interprets place recognition as a classification problem. We comprehensively evaluate our trained networks on several challenging benchmark place recognition datasets and demonstrate that they achieve an average 10% increase in performance over other place recognition algorithms and pre-trained CNNs. By analyzing the network responses and their differences from pre-trained networks, we provide insights into what a network learns when training for place recognition, and what these results signify for future research in this area.

Current self-driving car systems operate well in sunny weather but struggle in adverse conditions. One of the most commonly encountered adverse conditions involves water on the road caused by rain, sleet, melting snow or flooding. While some advances have been made in using conventional RGB camera and LIDAR technology for detecting water hazards, other sources of information such as polarization offer a promising and potentially superior approach to this problem in terms of performance and cost. In this paper, we present a novel stereo-polarization system for detecting and tracking water hazards based on polarization and color variation of reflected light, with consideration of the effect of polarized light from sky as function of reflection and azimuth angles. To evaluate this system, we present a new large `water on road' datasets spanning approximately 2 km of driving in various on-road and off-road conditions and demonstrate for the first time reliable water detection and tracking over a wide range of realistic car driving water conditions using polarized vision as the primary sensing modality. Our system successfully detects water hazards up to more than 100m. Finally, we discuss several interesting challenges and propose future research directions for further improving robust autonomous car perception in hazardous wet conditions using polarization sensors.

Deep learning models have achieved state-of-the-art performance in recognizing human activities, but often rely on utilizing background cues present in typical computer vision datasets that predominantly have a stationary camera. If these models are to be employed by autonomous robots in real world environments, they must be adapted to perform independently of background cues and camera motion effects. To address these challenges, we propose a new method that firstly generates generic action region proposals with good potential to locate one human action in unconstrained videos regardless of camera motion and then uses action proposals to extract and classify effective shape and motion features by a ConvNet framework. In a range of experiments, we demonstrate that by actively proposing action regions during both training and testing, state-of-the-art or better performance is achieved on benchmarks. We show the outperformance of our approach compared to the state-of-the-art in two new datasets; one emphasizes on irrelevant background, the other highlights the camera motion. We also validate our action recognition method in an abnormal behavior detection scenario to improve workplace safety. The results verify a higher success rate for our method due to the ability of our system to recognize human actions regardless of environment and camera motion.

The place recognition problem comprises two distinct subproblems; recognizing a specific location in the world (“specific” or “ordinary” place recognition) and recognizing the type of place (place categorization). Both are important competencies for mobile robots and have each received significant attention in the robotics and computer vision community, but usually as separate areas of investigation. In this paper, we leverage the powerful complementary nature of place recognition and place categorization processes to create a new hybrid place recognition system that uses place context to inform place recognition. We show that semantic place categorization creates an informative natural segmenting of physical space that in turn enables significantly better place recognition performance in comparison to existing techniques. In particular, this new semantically-informed approach adds robustness to significant local changes within the environment, such as transitioning between indoor and outdoor environments or between dark and light rooms in a house, complementing the capabilities of current condition-invariant techniques that are robust to globally consistent change (such as day to night cycles). We perform experiments using 4 novel benchmark datasets and show that semantically-informed place recognition outperforms the previous state-of-the-art systems. Like it does for object recognition [1], we believe that semantics can play a key role in boosting conventional place recognition and navigation performance for robotic systems.

Learning and recognition is a fundamental process performed in many robot operations such as mapping and localization. The majority of approaches share some common characteristics, such as attempting to extract salient features, landmarks or signatures, and growth in data storage and computational requirements as the size of the environment increases. In biological systems, spatial encoding in the brain is definitively known to be performed using a fixed-size neural encoding framework - the place, head-direction and grid cells found in the mammalian hippocampus and entorhinal cortex. Particularly paradoxically, one of the main encoding centers - the grid cells - represents the world using a highly aliased, repetitive encoding structure where one neuron represents an unbounded number of places in the world. Inspired by this system, in this paper we invert the normal approach used in forming mapping and localization algorithms, by developing a novel place recognition algorithm that seeks out and leverages repetitive, mutually complementary landmark frequencies in the world. The combinatorial encoding capacity of multiple different frequencies enables not only the ability to achieve efficient data storage, but also the potential for sub-linear storage growth in a learning and recall system. Using both ground-based and aerial camera datasets, we demonstrate the system finding and utilizing these frequencies to achieve successful place recognition, and discuss how this approach might scale to arbitrarily large global datasets and dimensions.

For intelligent robots to interact in meaningful ways with their environment, they must understand both the geometric and semantic properties of the scene surrounding them. The majority of research to date has addressed these mapping challenges separately, focusing on either geometric or semantic mapping. In this paper we address the problem of building environmental maps that include both semantically meaningful, object-level entities and point- or mesh-based geometrical representations. We simultaneously build geometric point cloud models of previously unseen instances of known object classes and create a map that contains these object models as central entities. Our system leverages sparse, feature-based RGB-D SLAM, image-based deep-learning object detection and 3D unsupervised segmentation.

Domestic service robots such as lawn mowing and vacuum cleaning robots are the most numerous consumer robots in existence today. While early versions employed random exploration, recent systems fielded by most of the major manufacturers have utilized range-based and visual sensors and user-placed beacons to enable robots to map and localize. However, active range and visual sensing solutions have the disadvantages of being intrusive, expensive, or only providing a 1D scan of the environment, while the requirement for beacon placement imposes other practical limitations. In this paper we present a passive and potentially cheap vision-based solution to 2D localization at night that combines easily obtainable day-time maps with low resolution contrast-normalized image matching algorithms, image sequence-based matching in two-dimensions, place match interpolation and recent advances in conventional low light camera technology. In a range of experiments over a domestic lawn and in a lounge room, we demonstrate that the proposed approach enables 2D localization at night, and analyse the effect on performance of varying odometry noise levels, place match interpolation and sequence matching length. Finally we benchmark the new low light camera technology and show how it can enable robust place recognition even in an environment lit only by a moonless sky, raising the tantalizing possibility of being able to apply all conventional vision algorithms, even in the darkest of nights.

Place recognition is a key capability for navigating robots. While significant advances have been achieved on large, stable platforms such as robot cars, achieving robust performance on rapidly manoeuvring platforms in outdoor natural conditions remains a challenge, with few systems able to deal with both variable conditions and large tilt variations caused by rough terrain. Taking inspiration from biology, we propose a novel combination of sensory modality and image processing to obtain a significant improvement in the robustness of sequence-based image matching for place recognition. We use a UV-sensitive fisheye lens camera to segment sky from ground, providing illumination invariance, and encode the resulting binary images using spherical harmonics to enable rotation-invariant image matching. In combination, these methods also produce substantial pitch and roll invariance, as the spherical harmonics for the sky shape are minimally affected, providing the sky remains visible. We evaluate the performance of our method against a leading appearance-invariant technique (SeqSLAM) and a leading viewpoint-invariant technique (FAB-MAP 2.0) on three new outdoor datasets encompassing variable robot heading, tilt, and lighting conditions in both forested and urban environments. The system demonstrates improved condition- and tilt-invariance, enabling robust place recognition during aggressive zigzag manoeuvring along bumpy trails and at tilt angles of up to 60 degrees.

In this paper we focus on the challenging problem of place categorization and semantic mapping on a robot without environment-specific training. Motivated by their ongoing success in various visual recognition tasks, we build our system upon a state-of-the-art convolutional network. We overcome its closed-set limitations by complementing the network with a series of one-vs-all classifiers that can learn to recognize new semantic classes online. Prior domain knowledge is incorporated by embedding the classification system into a Bayesian filter framework that also ensures temporal coherence. We evaluate the classification accuracy of the system on a robot that maps a variety of places on our campus in real-time. We show how semantic information can boost robotic object detection performance and how the semantic map can be used to modulate the robot's behaviour during navigation tasks. The system is made available to the community as a ROS module.

Robotic vision, unlike computer vision, typically involves processing a stream of images from a camera with time varying pose operating in an environment with time varying lighting conditions and moving objects. Repeating robotic vision experiments under identical conditions is often impossible, making it difficult to compare different algorithms. For machine learning applications a critical bottleneck is the limited amount of real world image data that can be captured and labelled for both training and testing purposes. In this paper we investigate the use of a photo-realistic simulation tool to address these challenges, in three specific domains: robust place recognition, visual SLAM and object recognition. For the first two problems we generate images from a complex 3D environment with systematically varying camera paths, camera viewpoints and lighting conditions. For the first time we are able to systematically characterise the performance of these algorithms as paths and lighting conditions change. In particular, we are able to systematically generate varying camera viewpoint datasets that would be difficult or impossible to generate in the real world. We also compare algorithm results for a camera in a real environment and a simulated camera in a simulation model of that real environment. Finally, for the object recognition domain, we generate labelled image data and characterise the viewpoint dependency of a current convolution neural network in performing object recognition. Together these results provide a multi-domain demonstration of the beneficial properties of using simulation to characterise and analyse a wide range of robotic vision algorithms.

This paper investigates the application of linear learning techniques to the place recognition problem. We present two learning methods, a supervised change prediction technique based on linear regression and an unsupervised change removal technique based on principal component analysis, and investigate how the performance of each is affected by the choice of training data. We show that the change prediction technique presented here succeeds only if it is provided with appropriate and adequate training data, which can be challenging for a mobile robotic system operating in an uncontrolled environment. In contrast, change removal can improve place recognition performance even when trained with as few as 100 samples. This paper shows that change removal can be combined with a number of different image descriptors and can improve performance across a range of different appearance conditions.

Visual place recognition is a challenging problem due to the vast range of ways in which the appearance of real-world places can vary. In recent years, improvements in visual sensing capabilities, an ever-increasing focus on long-term mobile robot autonomy, and the ability to draw on state-of-the-art research in other disciplines-particularly recognition in computer vision and animal navigation in neuroscience-have all contributed to significant advances in visual place recognition systems. This paper presents a survey of the visual place recognition research landscape. We start by introducing the concepts behind place recognition-the role of place recognition in the animal kingdom, how a “place” is defined in a robotics context, and the major components of a place recognition system. Long-term robot operations have revealed that changing appearance can be a significant factor in visual place recognition failure; therefore, we discuss how place recognition solutions can implicitly or explicitly account for appearance change within the environment. Finally, we close with a discussion on the future of visual place recognition, in particular with respect to the rapid advances being made in the related fields of deep learning, semantic scene understanding, and video description.

Robustness to variations in environmental conditions and camera viewpoint is essential for long-term place recognition, navigation and SLAM. Existing systems typically solve either of these problems, but invariance to both remains a challenge. This paper presents a training-free approach to lateral viewpoint- and condition-invariant, vision-based place recognition. Our successive frame patch-tracking technique infers average scene depth along traverses and automatically rescales views of the same place at different depths to increase their similarity. We combine our system with the condition-invariant SMART algorithm and demonstrate place recognition between day and night, across entire 4-lane-plus-median-strip roads, where current algorithms fail.

The recent focus on performing visual navigation and place recognition in changing environments has resulted in a large number of heterogeneous techniques each utilizing their own learnt or hand crafted visual features. This paper presents a generally applicable method for learning the appropriate distance metric by which to compare feature responses from any of these techniques in order to perform place recognition under changing environmental conditions. We implement an approach which learns to cluster images captured at spatially proximal locations under different conditions, separated from frames captured at different places. The formulation is a convex optimization, guaranteeing the existence of a global solution. We evaluate the general applicability of our method on two benchmark change datasets using three typical image pre-processing and feature types: GIST, Principal Component Analysis and learnt Convolutional Neural Network features. The results demonstrate that the distance metric learning approach uniformly improves single-image-based visual place recognition performance across all feature types. Furthermore, we demonstrate that this performance improvement is maintained when the sequence-based algorithm SeqSLAM is applied to the single-image place recognition results, leading to state-of-the-art performance.

Localization is a critical capability for many autonomous mobile robots. Implicit in all localization and place recognition techniques is the ability to decide whether a place is novel or familiar; the idea of “has this place been visited before?”. To make this decision accurately, localization systems typically require prior calibration or training on datasets from the specific type of operating environment; a requirement that is not practical in many environments or applications. One approach to this problem has been to remove the perfect data association requirement for place recognition systems by utilizing robust pose graphs. However, this approach requires a batch-based process and consequently does not solve the problem of real-time navigation in the interim before optimization. We present a novel technique which calibrates the localization system online as the robot moves through an environment, alleviating the need for prior calibration of a localization system. We leverage a robust pose graph optimization to determine a suitable model for our place recognition system which is sensor-invariant and does not require manually-labeled data for threshold tuning. The system continually refines place recognition parameters using sensory data and an optimized pose graph in order to increase the instantaneous precision of the location matches output by the underlying robust pose graph. We evaluate the system using the public New College dataset and a multi-sensor indoor office dataset. Our results indicate that the system produces superior mapping performance to a static place recognition model, and superior instantaneous place recognition performance to a robust pose graph technique.

This paper is concerned with the interpretation of visual information for robot localization. It presents a probabilistic localization system that generates an appropriate observation model online, unlike existing systems which require pre-determined belief models. This paper proposes that probabilistic visual localization requires two major operating modes - one to match locations under similar conditions and the other to match locations under different conditions. We develop dual observation likelihood models to suit these two different states, along with a similarity measure-based method that identifies the current conditions and switches between the models. The system is experimentally tested against different types of ongoing appearance change. The results demonstrate that the system is compatible with a wide range of visual front-ends, and the dual-model system outperforms a single-model or pre-trained approach and state-of-the-art localization techniques.

After the incredible success of deep learning in the computer vision domain, there has been much interest in applying Convolutional Network (ConvNet) features in robotic fields such as visual navigation and SLAM. Unfortunately, there are fundamental differences and challenges involved. Computer vision datasets are very different in character to robotic camera data, real-time performance is essential, and performance priorities can be different. This paper comprehensively evaluates and compares the utility of three state-of-the-art ConvNets on the problems of particular relevance to navigation for robots; viewpoint-invariance and condition-invariance, and for the first time enables real-time place recognition performance using ConvNets with large maps by integrating a variety of existing (locality-sensitive hashing) and novel (semantic search space partitioning) optimization techniques. We present extensive experiments on four real world datasets cultivated to evaluate each of the specific challenges in place recognition. The results demonstrate that speed-ups of two orders of magnitude can be achieved with minimal accuracy degradation, enabling real-time performance. We confirm that networks trained for semantic place categorization also perform better at (specific) place recognition when faced with severe appearance changes and provide a reference for which networks and layers are optimal for different aspects of the place recognition problem.

This paper presents a novel place recognition algorithm inspired by the recent discovery of overlapping and multi-scale spatial maps in the rodent brain. We mimic this hierarchical framework by training arrays of Support Vector Machines to recognize places at multiple spatial scales. Place match hypotheses are then cross-validated across all spatial scales, a process which combines the spatial specificity of the finest spatial map with the consensus provided by broader mapping scales. Experiments on three real-world datasets including a large robotics benchmark demonstrate that mapping over multiple scales uniformly improves place recognition performance over a single scale approach without sacrificing localization accuracy. We present analysis that illustrates how matching over multiple scales leads to better place recognition performance and discuss several promising areas for future investigation.

Visual localization in outdoor environments is often hampered by the natural variation in appearance caused by such things as weather phenomena, diurnal fluctuations in lighting, and seasonal changes. Such changes are global across an environment and, in the case of global light changes and seasonal variation, the change in appearance occurs in a regular, cyclic manner. Visual localization could be greatly improved if it were possible to predict the appearance of a particular location at a particular time, based on the appearance of the location in the past and knowledge of the nature of appearance change over time. In this paper, we investigate whether global appearance changes in an environment can be learned sufficiently to improve visual localization performance. We use time of day as a test case, and generate transformations between morning and afternoon using sample images from a training set. We demonstrate the learned transformation can be generalized from training data and show the resulting visual localization on a test set is improved relative to raw image comparison. The improvement in localization remains when the area is revisited several weeks later.

Whole image descriptors have been shown to be remarkably robust to perceptual change especially compared to local features. However, whole-image-based localization systems typically rely on heuristic methods for determining appropriate matching thresholds in a particular environment. These environment-specific tuning requirements and the lack of a meaningful interpretation of arbitrary thresholds limit the general applicability of these systems. In this paper we present a Bayesian model of probability for whole-image descriptors that can be seamlessly integrated into localization systems designed for probabilistic visual input. We demonstrate this method using CAT-Graph, an appearance-based visual localization system originally designed for a FAB-MAP-style probabilistic input. We show that using whole-image descriptors as visual input extends CAT-Graph's functionality to environments that experience a greater amount of perceptual change. We also present a method of estimating whole-image probability models in an online manner, removing the need for a prior training phase. We show that this online, automated training method can perform comparably to pre-trained, manually tuned local descriptor methods.

In this paper we present a novel, condition-invariant place recognition algorithm inspired by recent discoveries in human visual neuroscience. The algorithm combines intolerant but fast low resolution whole image matching with highly tolerant, sub-image patch matching processes. The approach does not require prior training and works on single images, alleviating the need for either a velocity signal or image sequence, differentiating it from current state of the art methods. We conduct an exhaustive set of experiments evaluating the relationship between place recognition performance and computational resources using part of the challenging Alderley sunny day - rainy night dataset, which has only been previously solved by integrating over 320 frame long image sequences. We achieve recall rates of up to 51% at 100% precision, matching places that have undergone drastic perceptual change while rejecting match hypotheses between highly aliased images of different places. Human trials demonstrate the performance is approaching human capability. The results provide a new benchmark for single image, condition-invariant place recognition.

This paper presents Sequence Matching Across Route Traversals (SMART); a generally applicable sequence-based place recognition algorithm. SMART provides invariance to changes in illumination and vehicle speed while also providing moderate pose invariance and robustness to environmental aliasing. We evaluate SMART on vehicles travelling at highly variable speeds in two challenging environments; firstly, on an all-terrain vehicle in an off-road, forest track and secondly, using a passenger car traversing an urban environment across day and night. We provide comparative results to the current state-of-the-art SeqSLAM algorithm and investigate the effects of altering SMART's image matching parameters. Additionally, we conduct an extensive study of the relationship between image sequence length and SMART's matching performance. Our results show viable place recognition performance in both environments with short 10-metre sequences, and up to 96% recall at 100% precision across extreme day-night cycles when longer image sequences are used.

Wi-Fi is a commonly available source of localization information in urban environments but is challenging to integrate into conventional mapping architectures. Current state of the art probabilistic Wi-Fi SLAM algorithms are limited by spatial resolution and an inability to remove the accumulation of rotational error, inherent limitations of the Wi-Fi architecture. In this paper we leverage the low quality sensory requirements and coarse metric properties of RatSLAM to localize using Wi-Fi fingerprints. To further improve performance, we present a novel sensor fusion technique that integrates camera and Wi-Fi to improve localization specificity, and use compass sensor data to remove orientation drift. We evaluate the algorithms in diverse real world indoor and outdoor environments, including an office floor, university campus and a visually aliased circular building loop. The algorithms produce topologically correct maps that are superior to those produced using only a single sensor modality.

Current state of the art robot mapping and navigation systems produce impressive performance under a narrow range of robot platform, sensor and environmental conditions, in contrast to animals such as rats that produce “good enough” maps that enable them to function under an incredible range of situations. In this paper we present a rat-inspired featureless sensor-fusion system that assesses the usefulness of multiple sensor modalities based on their utility and coherence for place recognition, without knowledge as to the type of sensor. We demonstrate the system on a Pioneer robot in indoor and outdoor environments with abrupt lighting changes. Through dynamic weighting of the sensors, the system is able to perform correct place recognition and mapping where the static sensor weighting approach fails.

In this paper we demonstrate passive vision-based localization in environments more than two orders of magnitude darker than the current benchmark using a $100 webcam and a $500 camera. Our approach uses the camera's maximum exposure duration and sensor gain to achieve appropriately exposed images even in unlit night-time environments, albeit with extreme levels of motion blur. Using the SeqSLAM algorithm, we first evaluate the effect of variable motion blur caused by simulated exposures of 132 ms to 10000 ms duration on localization performance. We then use actual long exposure camera datasets to demonstrate day-night localization in two different environments. Finally we perform a statistical analysis that compares the baseline performance of matching unprocessed grayscale images to using patch normalization and local neighborhood normalization - the two key SeqSLAM components. Our results and analysis show for the first time why the SeqSLAM algorithm is effective, and demonstrate the potential for cheap camera-based localization systems that function across extreme perceptual change.

In this paper we present a method for autonomously tuning the threshold between learning and recognizing a place in the world, based on both how the rodent brain is thought to process and calibrate multisensory data and the pivoting movement behaviour that rodents perform in doing so. The approach makes no assumptions about the number and type of sensors, the robot platform, or the environment, relying only on the ability of a robot to perform two revolutions on the spot. In addition, it self-assesses the quality of the tuning process in order to identify situations in which tuning may have failed. We demonstrate the autonomous movement-driven threshold tuning on a Pioneer 3DX robot in eight locations spread over an office environment and a building car park, and then evaluate the mapping capability of the system on journeys through these environments. The system is able to pick a place recognition threshold that enables successful environment mapping in six of the eight locations while also autonomously flagging the tuning failure in the remaining two locations. We discuss how the method, in combination with parallel work on autonomous weighting of individual sensors, moves the parameter dependent RatSLAM system significantly closer to sensor, platform and environment agnostic operation.

A major challenge for robot localization and mapping systems is maintaining reliable operation in a changing environment. Vision-based systems in particular are susceptible to changes in illumination and weather, and the same location at another time of day may appear radically different to a system using a feature-based visual localization system. One approach for mapping changing environments is to create and maintain maps that contain multiple representations of each physical location in a topological framework or manifold. However, this requires the system to be able to correctly link two or more appearance representations to the same spatial location, even though the representations may appear quite dissimilar. This paper proposes a method of linking visual representations from the same location without requiring a visual match, thereby allowing vision-based localization systems to create multiple appearance representations of physical locations. The most likely position on the robot path is determined using particle filter methods based on dead reckoning data and recent visual loop closures. In order to avoid erroneous loop closures, the odometry-based inferences are only accepted when the inferred path's end point is confirmed as correct by the visual matching system. Algorithm performance is demonstrated using an indoor robot dataset and a large outdoor camera dataset.

Appearance-based loop closure techniques, which leverage the high information content of visual images and can be used independently of pose, are now widely used in robotic applications. The current state-of-the-art in the field is Fast Appearance-Based Mapping (FAB-MAP) having been demonstrated in several seminal robotic mapping experiments. In this paper, we describe OpenFABMAP, a fully open source implementation of the original FAB-MAP algorithm. Beyond the benefits of full user access to the source code, OpenFABMAP provides a number of configurable options including rapid codebook training and interest point feature tuning. We demonstrate the performance of OpenFABMAP on a number of published datasets and demonstrate the advantages of quick algorithm customisation. We present results from OpenFABMAP's application in a highly varied range of robotics research scenarios.

Appearance-based localization is increasingly used for loop closure detection in metric SLAM systems. Since it relies only upon the appearance-based similarity between images from two locations, it can perform loop closure regardless of accumulated metric error. However, the computation time and memory requirements of current appearance-based methods scale linearly not only with the size of the environment but also with the operation time of the platform. These properties impose severe restrictions on longterm autonomy for mobile robots, as loop closure performance will inevitably degrade with increased operation time. We present a set of improvements to the appearance-based SLAM algorithm CAT-SLAM to constrain computation scaling and memory usage with minimal degradation in performance over time. The appearance-based comparison stage is accelerated by exploiting properties of the particle observation update, and nodes in the continuous trajectory map are removed according to minimal information loss criteria. We demonstrate constant time and space loop closure detection in a large urban environment with recall performance exceeding FAB-MAP by a factor of 3 at 100% precision, and investigate the minimum computational and memory requirements for maintaining mapping performance.

Learning and then recognizing a route, whether travelled during the day or at night, in clear or inclement weather, and in summer or winter is a challenging task for state of the art algorithms in computer vision and robotics. In this paper, we present a new approach to visual navigation under changing conditions dubbed SeqSLAM. Instead of calculating the single location most likely given a current image, our approach calculates the best candidate matching location within every local navigation sequence. Localization is then achieved by recognizing coherent sequences of these “local best matches”. This approach removes the need for global matching performance by the vision front-end - instead it must only pick the best match within any short sequence of images. The approach is applicable over environment changes that render traditional feature-based techniques ineffective. Using two car-mounted camera datasets we demonstrate the effectiveness of the algorithm and compare it to one of the most successful feature-based SLAM algorithms, FAB-MAP. The perceptual change in the datasets is extreme; repeated traverses through environments during the day and then in the middle of the night, at times separated by months or years and in opposite seasons, and in clear weather and extremely heavy rain. While the feature-based method fails, the sequence-based algorithm is able to match trajectory segments at 100% precision with recall rates of up to 60%.

The challenge of persistent appearance-based navigation and mapping is to develop an autonomous robotic vision system that can simultaneously localize, map and navigate over the lifetime of the robot. However, the computation time and memory requirements of current appearance-based methods typically scale not only with the size of the environment but also with the operation time of the platform; also, repeated revisits to locations will develop multiple competing representations which reduce recall performance. In this paper we present a solution to the persistent localization, mapping and global path planning problem in the context of a delivery robot in an office environment over a one-week period. Using a graphical appearance-based SLAM algorithm, CAT-Graph, we demonstrate constant time and memory loop closure detection with minimal degradation during repeated revisits to locations, along with topological path planning that improves over time without using a global metric representation. We compare the localization performance of CAT-Graph to openFABMAP, an appearance-only SLAM algorithm, and the path planning performance to occupancy-grid based metric SLAM. We discuss the limitations of the algorithm with regard to environment change over time and illustrate how the topological graph representation can be coupled with local movement behaviors for persistent autonomous robot navigation.

Spatial navigation requires the processing of complex, disparate and often ambiguous sensory data. The neurocomputations underpinning this vital ability remain poorly understood. Controversy remains as to whether multimodal sensory information must be combined into a unified representation, consistent with Tolman's "cognitive map", or whether differential activation of independent navigation modules suffice to explain observed navigation behaviour. Here we demonstrate that key neural correlates of spatial navigation in darkness cannot be explained if the path integration system acted independently of boundary (landmark) information. In vivo recordings demonstrate that the rodent head direction (HD) system becomes unstable within three minutes without vision. In contrast, rodents maintain stable place fields and grid fields for over half an hour without vision. Using a simple HD error model, we show analytically that idiothetic path integration (iPI) alone cannot be used to maintain any stable place representation beyond two to three minutes. We then use a measure of place stability based on information theoretic principles to prove that featureless boundaries alone cannot be used to improve localization above chance level. Having shown that neither iPI nor boundaries alone are sufficient, we then address the question of whether their combination is sufficient and--we conjecture--necessary to maintain place stability for prolonged periods without vision. We addressed this question in simulations and robot experiments using a navigation model comprising of a particle filter and boundary map. The model replicates published experimental results on place field and grid field stability without vision, and makes testable predictions including place field splitting and grid field rescaling if the true arena geometry differs from the acquired boundary map. We discuss our findings in light of current theories of animal navigation and neuronal computation, and elaborate on their implications and significance for the design, analysis and interpretation of experiments.

This paper describes a novel probabilistic approach to incorporating odometric information into appearance-based SLAM systems, without performing metric map construction or calculating relative feature geometry. The proposed system, dubbed Continuous Appearance-based Trajectory SLAM (CAT-SLAM), represents location as a probability distribution along a trajectory, and represents appearance continuously over the trajectory rather than at discrete locations. The distribution is evaluated using a Rao Blackwellised particle filter, which weights particles based on local appearance and odometric similarity and explicitly models both the likelihood of revisiting previous locations and visiting new locations. A modified resampling scheme counters particle deprivation and allows loop closure updates to be performed in constant time regardless of map size. We compare the performance of CAT-SLAM to FAB-MAP (an appearance-only SLAM algorithm) in an outdoor environment, demonstrating a threefold increase in the number of correct loop closures detected by CAT-SLAM.

Micro aerial vehicles (MAVs) are a rapidly growing area of research and development in robotics. For autonomous robot operations, localization has typically been calculated using GPS, external camera arrays, or onboard range or vision sensing. In cluttered indoor or outdoor environments, onboard sensing is the only viable option. In this paper we present an appearance-based approach to visual SLAM on a flying MAV using only low quality vision. Our approach consists of a visual place recognition algorithm that operates on 1000 pixel images, a lightweight visual odometry algorithm, and a visual expectation algorithm that improves the recall of place sequences and the precision with which they are recalled as the robot flies along a similar path. Using data gathered from outdoor datasets, we show that the system is able to perform visual recognition with low quality, intermittent visual sensory data. By combining the visual algorithms with the RatSLAM system, we also demonstrate how the algorithms enable successful SLAM.

The Lingodroids are a pair of mobile robots that evolve a language for places and relationships between places (based on distance and direction). Each robot in these studies has its own understanding of the layout of the world, based on its unique experiences and exploration of the environment. Despite having different internal representations of the world, the robots are able to develop a common lexicon for places, and then use simple sentences to explain and understand relationships between places even places that they could not physically experience, such as areas behind closed doors. By learning the language, the robots are able to develop representations for places that are inaccessible to them, and later, when the doors are opened, use those representations to perform goal-directed behavior.

Appearance-based mapping and localisation is especially challenging when separate processes of mapping and localisation occur at different times of day. The problem is exacerbated in the outdoors where continuous change in sun angle can drastically affect the appearance of a scene. We confront this challenge by fusing the probabilistic local feature based data association method of FAB-MAP with the pose cell filtering and experience mapping of RatSLAM. We evaluate the effectiveness of our amalgamation of methods using five datasets captured throughout the day from a single camera driven through a network of suburban streets. We show further results when the streets are re-visited three weeks later, and draw conclusions on the value of the system for lifelong mapping.

To successfully navigate their habitats, many mammals use a combination of two mechanisms, path integration and calibration using landmarks, which together enable them to estimate their location and orientation, or pose. In large natural environments, both these mechanisms are characterized by uncertainty: the path integration process is subject to the accumulation of error, while landmark calibration is limited by perceptual ambiguity. It remains unclear how animals form coherent spatial representations in the presence of such uncertainty. Navigation research using robots has determined that uncertainty can be effectively addressed by maintaining multiple probabilistic estimates of a robot's pose. Here we show how conjunctive grid cells in dorsocaudal medial entorhinal cortex (dMEC) may maintain multiple estimates of pose using a brain-based robot navigation system known as RatSLAM. Based both on rodent spatially-responsive cells and functional engineering principles, the cells at the core of the RatSLAM computational model have similar characteristics to rodent grid cells, which we demonstrate by replicating the seminal Moser experiments. We apply the RatSLAM model to a new experimental paradigm designed to examine the responses of a robot or animal in the presence of perceptual ambiguity. Our computational approach enables us to observe short-term population coding of multiple location hypotheses, a phenomenon which would not be easily observable in rodent recordings. We present behavioral and neural evidence demonstrating that the conjunctive grid cells maintain and propagate multiple estimates of pose, enabling the correct pose estimate to be resolved over time even without uniquely identifying cues. While recent research has focused on the grid-like firing characteristics, accuracy and representational capacity of grid cells, our results identify a possible critical and unique role for conjunctive grid cells in filtering sensory uncertainty. We anticipate our study to be a starting point for animal experiments that test navigation in perceptually ambiguous environments.

The implementation of a robotic security solution generally requires one algorithm to route the robot around the environment and another algorithm to perform anomaly detection. Solutions to the routing problem require the robot to have a good estimate of its own pose. We present a novel security system that uses metrics generated by the localisation algorithm to perform adaptive anomaly detection. The localisation algorithm is a vision-based SLAM solution called RatSLAM, based on mechanisms within the hippocampus. The anomaly detection algorithm is based on the mechanisms used by the immune system to identify threats to the body. The system is explored using data gathered within an unmodified office environment. It is shown that the algorithm successfully reacts to the presence of people and objects in areas where they are not usually present and is tolerised against the presence of people in environments that are usually dynamic.

Simultaneous localization and mapping (SLAM) is one of the major challenges in mobile robotics. Probabilistic techniques using high-end range finding devices are well established in the field, but work has investigated vision-only approaches. This paper presents a method for generating approximate rotational and translation velocity information from a single vehicle-mounted consumer camera, without the computationally expensive process of tracking landmarks. The method is tested by employing it to provide the odometric and visual information for the RatSLAM system while mapping a complex suburban road network. RatSLAM generates a coherent map of the environment during an 18 km long trip through suburban traffic at speeds of up to 60 km/hr. This result demonstrates the potential of ground-based vision-only SLAM using low cost sensing and computational hardware.

This paper describes a biologically inspired approach to vision-only simultaneous localization and mapping (SLAM) on ground-based platforms. The core SLAM system, dubbed RatSLAM, is based on computational models of the rodent hippocampus, and is coupled with a lightweight vision system that provides odometry and appearance information. RatSLAM builds a map in an online manner, driving loop closure and relocalization through sequences of familiar visual scenes. Visual ambiguity is managed by maintaining multiple competing vehicle pose estimates, while cumulative errors in odometry are corrected after loop closure by a map correction algorithm. We demonstrate the mapping performance of the system on a 66 km car journey through a complex suburban road network. Using only a web camera operating at 10 Hz, RatSLAM generates a coherent map of the entire environment at real-time speed, correctly closing more than 51 loops of up to 5 km in length.

The RatSLAM system can perform vision based SLAM using a computational model of the rodent hippocampus. When the number of pose cells used to represent space in RatSLAM is reduced, artifacts are introduced that hinder its use for goal directed navigation. This paper describes a new component for the RatSLAM system called an experience map, which provides a coherent representation for goal directed navigation. Results are presented for two sets of real world experiments, including comparison with the original goal memory system's performance in the same environment. Preliminary results are also presented demonstrating the ability of the experience map to adapt to simple short term changes in the environment

RatSLAM is a system for vision based Simultaneous Localization and Mapping (SLAM) that has been shown to be capable of building stable representations of real world environments. In this paper we describe a method for using RatSLAM representations as the basis for navigation to designated goal locations. The method uses a new component, goal memory, to learn the temporal gradient between places. Paths are recalled or inferred from the goal memory by following the temporal gradient from the robot’s current position to the goal location. Experimental results have been gathered in a combined office and laboratory environment using a Pioneer robot. The experiments show that the robot can perform vision based SLAM on-line and in real time, and then use those representations immediately to navigate directly to designated goal locations.

The work presents a new approach to the problem of simultaneous localization and mapping - SLAM - inspired by computational models of the hippocampus of rodents. The rodent hippocampus has been extensively studied with respect to navigation tasks, and displays many of the properties of a desirable SLAM solution. RatSLAM is an implementation of a hippocampal model that can perform SLAM in real time on a real robot. It uses a competitive attractor network to integrate odometric information with landmark sensing to form a consistent representation of the environment. Experimental results show that RatSLAM can operate with ambiguous landmark information and recover from both minor and major path integration errors.

This paper illustrates a method for finding useful visual landmarks for performing simultaneous localization and mapping (SLAM). The method is based loosely on biological principles, using layers of filtering and pooling to create learned templates that correspond to different views of the environment. Rather than using a set of landmarks and reporting range and bearing to the landmark, this system maps views to poses. The challenge is to produce a system that produces the same view for small changes in robot pose, but provides different views for larger changes in pose. The method has been developed to interface with the RatSLAM system, a biologically inspired method of SLAM. The paper describes the method of learning and recalling visual landmarks in detail, and shows the performance of the visual system in real robot tests.