A central challenge in building continually improving agents is that training environments are typically static or manually constructed. This restricts continual learning and generalization beyond the training distribution. We address this with COvolve, a co-evolutionary framework that leverages large language models (LLMs) to generate both environments and agent policies, expressed as executable Python code. We model the interaction between environment and policy designers as a two-player zero-sum game, ensuring adversarial co-evolution in which environments expose policy weaknesses and policies adapt in response. This process induces an automated curriculum in which environments and policies co-evolve toward increasing complexity. To guarantee robustness and prevent forgetting as the curriculum progresses, we compute the mixed-strategy Nash equilibrium (MSNE) of the zero-sum game, thereby yielding a meta-policy. This MSNE meta-policy ensures that the agent does not forget to solve previously seen environments while learning to solve previously unseen ones. Experiments in urban driving, symbolic maze-solving, and geometric navigation showcase that COvolve produces progressively more complex environments. Our results demonstrate the potential of LLM-driven co-evolution to achieve open-ended learning without predefined task distributions or manual intervention.

Human-induced errors—such as slips, lapses and mistakes—are natural and often unavoidable, posing significant safety risks in teleoperation, particularly in high-risk, dynamic environments like underwater operations. While previous work has emphasised system-level enhancements, the proactive mitigation of human-induced errors through empirical evaluation remains underexplored. This work presents the first empirical investigation of Intelligent Disobedience (ID) as a collaborative strategy for mitigating human errors in teleoperation. A semi-controlled, game-based Wizard-of-Oz study with 40 participants performing an underwater navigation task was conducted. Two ID strategies were evaluated: one with automatic robot mitigation and another with user-mediated robot mitigation following the disobedient intervention. Using a mixed-methods approach that triangulated quantitative performance metrics with qualitative insights from semi-structured interviews, the study examined the effects of these strategies on operator performance, acceptance, and error-mitigation outcomes across different error types. The results revealed no detectable performance degradation under ID and emphasised that human error types differ in their characteristics, necessitating the adaptation of ID intervention strategies. These findings underscore the importance of context-adaptive approaches when integrating ID into collaborative teleoperation systems.

Designing effective reward functions is crucial to training reinforcement learning (RL) algorithms. However, this design is non-trivial, even for domain experts, due to the subjective nature of certain tasks that are hard to quantify explicitly. In recent works, large language models (LLMs) have been used for reward generation from natural language task descriptions, leveraging their extensive instruction tuning and commonsense understanding of human behavior. In this work, we hypothesize that LLMs, guided by human feedback, can be used to formulate reward functions that reflect human implicit knowledge. We study this in three challenging settings -- autonomous driving, humanoid locomotion, and dexterous manipulation -- wherein notions of ``good" behavior are tacit and hard to quantify. To this end, we introduce REvolve, a truly evolutionary framework that uses LLMs for reward design in RL. REvolve generates and refines reward functions by utilizing human feedback to guide the evolution process, effectively translating implicit human knowledge into explicit reward functions for training (deep) RL agents. Experimentally, we demonstrate that agents trained on REvolve-designed rewards outperform other state-of-the-art baselines.

This study investigates the efficiency and safety outcomes of implementing different adaptive coordination models for automated vehicle (AV) fleets, managed by a centralized coordinator that dynamically responds to human-controlled vehicle behavior. The simulated scenarios replicate an underground mining environment characterized by narrow tunnels with limited connectivity. To address the unique challenges of such settings, we propose a novel metric - Path Overlap Density (POD)- to predict efficiency and potentially the safety performance of AV fleets. The study also explores the impact of map features on AV fleets performance. The results demonstrate that both AV fleet coordination strategies and underground tunnel network characteristics significantly influence overall system performance. While map features are critical for optimizing efficiency, adaptive coordination strategies are essential for ensuring safe operations.

Integrating human expertise with machine learning is crucial for applications demanding high accuracy and safety, such as autonomous driving. This study introduces Interactive Double Deep Q-network (iDDQN), a Human-in-the-Loop (HITL) approach that enhances Reinforcement Learning (RL) by merging human insights directly into the RL training process, improving model performance. Our proposed iDDQN method modifies the Q-value update equation to integrate human and agent actions, establishing a collaborative approach for policy development. Additionally, we present an offline evaluative framework that simulates the agent's trajectory as if no human intervention to assess the effectiveness of human interventions. Empirical results in simulated autonomous driving scenarios demonstrate that iDDQN outperforms established approaches, including Behavioral Cloning (BC), HG-DAgger, Deep Q-Learning from Demonstrations (DQfD), and vanilla DRL in leveraging human expertise for improving performance and adaptability.

This work highlights the challenge of labeling data with single-label categories, as there may be ambiguity in the assigned labels. This ambiguity arises when a data sample, which can be influenced by previous affective events is labeled with a single-label category (known as priming). Label distribution learning (LDL) is proposed as an approach to contend with the ambiguity among labels. This approach has been relatively unexplored in the field of affective computing. In this work, an experiment is designed to explore the benefits of employing LDL using specifically the SEED and SEED-V datasets. In these datasets, different emotions are induced by exposing participants to a sequence of stimuli (videoclip watching). However, these datasets provide single labels, where each data point corresponds to one affective state or emotion. Due to the lack of label distributions within existing benchmarks, label enhancement serves as a preparatory step, whose goal is to compute label distributions from the feature space and single labels before training a label distribution learning model. Experimental results show that the LDL approach reduces confusion with respect to the emotion induced in the previous trial. Distribution learning is an approach that can help to further improve the prediction of affect, which to date remains a difficult and ambiguous concept to label.

Transformer-based models have gained increasing popularity for time-series prediction; however, in specific applications such as residential heating systems, static contextual data of buildings is crucial to effectively capture and learn complex environmental dynamics. This paper presents a novel transformer-based model that adapts the contextual meta-data of residential buildings, generalizing across diverse environments. The model integrates temporal data with adaptive embedding of building-specific contextual meta-data such as geographic locations and building characteristics to dynamically learn and adapt to the variations. These adaptive context embeddings allow the model to comprehensively understand how different buildings respond to environmental changes over time. Initial results show improved accuracy and reliability in indoor temperature predictions of residential buildings, demonstrating the model’s potential to optimize district heating systems across a diverse array of residential buildings. This proposed model provides a basis for developing proactive heat management systems in buildings.

This article studies group-wise point set registration and makes the following contributions: "FuzzyGReg", which is a new fuzzy cluster-based method to register multiple point sets jointly, and "FuzzyQA", which is the associated quality assessment to check registration accuracy automatically. Given a group of point sets, FuzzyGReg creates a model of fuzzy clusters and equally treats all the point sets as the elements of the fuzzy clusters. Then, the group-wise registration is turned into a fuzzy clustering problem. To resolve this problem, FuzzyGReg applies a fuzzy clustering algorithm to identify the parameters of the fuzzy clusters while jointly transforming all the point sets to achieve an alignment. Next, based on the identified fuzzy clusters, FuzzyQA calculates the spatial properties of the transformed point sets and then checks the alignment accuracy by comparing the similarity degrees of the spatial properties of the point sets. When a local misalignment is detected, a local re-alignment is performed to improve accuracy. The proposed method is cost-efficient and convenient to be implemented. In addition, it provides reliable quality assessments in the absence of ground truth and user intervention. In the experiments, different point sets are used to test the proposed method and make comparisons with state-of-the-art registration techniques. The experimental results demonstrate the effectiveness of our method. The code is available at https://gitsvn-nt.oru.se/qianfang.liao/FuzzyGRegWithQA.

Q-Networks are used in Reinforcement Learning to model the expected return from every action at a given state. When training Q-Networks, external reward signals are propagated to the previously performed actions leading up to each reward. If many actions are required before experiencing a reward, the reward signal is distributed across all those actions, where some actions may have greater impact on the reward than others. As the number of significant actions between rewards increases, the relative importance of each action decreases. If actions have too small importance, their impact might be overshadowed by noise in a deep neural network model, potentially causing convergence issues. In this work, we empirically test the limits of increasing the number of actions leading up to a reward in a simple grid-world environment. We show in our experiments that even though the training error surpasses the reward signal attributed to each action, the model is still able to learn a smooth enough value representation.

Safety in human-robot interaction can be divided into physical safety and perceived safety, where the latter is still under-addressed in the literature. Investigating perceived safety in human-robot interaction requires a multidisciplinary perspective. Indeed, perceived safety is often considered as being associated with several common factors studied in other disciplines, i.e., comfort, predictability, sense of control, and trust. In this paper, we investigated the relationship between these factors and perceived safety in human-robot interaction using subjective and objective measures. We conducted a two-by-five mixed-subjects design experiment. There were two between-subjects conditions: the faulty robot was experienced at the beginning or the end of the interaction. The five within-subjects conditions correspond to (1) baseline, and the manipulations of robot behaviors to stimulate: (2) discomfort, (3) decreased perceived safety, (4) decreased sense of control and (5) distrust. The idea of triggering a deprivation of these factors was motivated by the definition of safety in the literature where safety is often defined by the absence of it. Twenty-seven young adult participants took part in the experiments. Participants were asked to answer questionnaires that measure the manipulated factors after within-subjects conditions. Besides questionnaire data, we collected objective measures such as videos and physiological data. The questionnaire results show a correlation between comfort, sense of control, trust, and perceived safety. Since these factors are the main factors that influence perceived safety, they should be considered in human-robot interaction design decisions. We also discuss the effect of individual human characteristics (such as personality and gender) that they could be predictors of perceived safety. We used the physiological signal data and facial affect from videos for estimating perceived safety where participants’ subjective ratings were utilized as labels. The data from objective measures revealed that the prediction rate was higher from physiological signal data. This paper can play an important role in the goal of better understanding perceived safety in human-robot interaction.

Caregivers in eldercare can benefit from telepresence robots that allow them to perform a variety of tasks remotely. In order for such robots to be operated effectively and efficiently by non-technical users, it is important to examine if and how the robotic system’s level of automation (LOA) impacts their performance. The objective of this work was to develop suitable LOA modes for a mobile robotic telepresence (MRP) system for eldercare and assess their influence on users’ performance, workload, awareness of the environment, and usability at two different levels of task complexity. For this purpose, two LOA modes were implemented on the MRP platform: assisted teleoperation (low LOA mode) and autonomous navigation (high LOA mode). The system was evaluated in a user study with 20 participants, who, in the role of the caregiver, navigated the robot through a home-like environment to perform control and perception tasks. Results revealed that performance improved in the high LOA when task complexity was low. However, when task complexity increased, lower LOA improved performance. This opposite trend was also observed in the results for workload and situation awareness. We discuss the results in terms of the LOAs’ impact on users’ attitude towards automation and implications on usability.

Imitation learning is an important direction in the area of robot skill learning. It provides a user-friendly and straightforward solution to transfer human demonstrations to robots. In this article, we integrate fuzzy theory into imitation learning to develop a novel method called type-2 fuzzy model-based movement primitives (T2FMP). In this method, a group of data-driven type-2 fuzzy models are used to describe the input–output relationships of demonstrations. Based on the fuzzy models, T2FMP can efficiently reproduce the trajectory without high computational costs or cumbersome parameter settings. Besides, it can well handle the variation of the demonstrations and is robust to noise. In addition, we develop extensions that endow T2FMP with trajectory modulation and superposition to achieve real-time trajectory adaptation to various scenarios. Going beyond existing imitation learning methods, we further extend T2FMP to regulate the trajectory to avoid collisions in the environment that is unstructured, nonconvex, and detected with noisy outliers. Several experiments are performed to validate the effectiveness of our method.

Large–scale visual sensor networks have become progressively an essential part of our daily lives underpinning many technological, financial, and social advancements today, with applications in smart cities, traffic monitoring, environmental pollution control, public safety, and crime prevention.

Modeling object representations derived from perceptual observations, in a way that is also semantically meaningful for humans as well as autonomous agents, is a prerequisite for joint human-agent understanding of the world. A practical approach that aims to model such representations is perceptual anchoring, which handles the problem of mapping sub-symbolic sensor data to symbols and maintains these mappings over time. In this paper, we present ProbAnch, a modular data-driven anchoring framework, whose implementation requires a variety of well-orchestrated components, including a probabilistic reasoning system.

Virtual reality (VR) is regarded as a useful tool for teleoperation systems and provides operators with immersive visual feedback on the robot and the environment. However, without any haptic feedback or physical constructions, VR-based teleoperation systems normally suffer from poor maneuverability, and operational faults may be caused in some fine movements. In this article, we employ mixed reality (MR), which combines real and virtual worlds, to develop a novel teleoperation system. A new system design and control algorithms are proposed. For the system design, an MR interface is developed based on a virtual environment augmented with real-time data from the task space with the goal of enhancing the operator's visual perception. To allow the operator to be freely decoupled from the control loop and offload the operator's burden, a new interaction proxy is proposed to control the robot. For the control algorithms, two control modes are introduced to improve the long-distance movements and fine movements of the MR-based teleoperation system. In addition, a set of fuzzy-logic-based methods are proposed to regulate the orientation, position, velocity, and force of the robot to enhance the system's maneuverability and address potential operational faults. A barrier Lyapunov function and a backstepping method are leveraged to design the control laws and simultaneously guarantee the system's stability under state constraints. Experiments conducted using a six-degree-of-freedom robotic arm prove the feasibility of the system.

This article proposes a new single master bimanual teleoperation system with an efficient position, orientation, and force regulation strategy. Unlike many existing studies that solely support motion synchronization, the first contribution of the proposed work is to propose a solution for orientation regulation when several slave robots have differing motions. In other words, we propose a solution for self-regulated orientation for dual-arm robots. A second contribution in the article allows the master with fewer degrees of freedom (DoF) to control the slaves (with higher DoF), while the orientation of the slaves is self-regulated. The system further offers a novel force regulation that enables the slave robots to have a smooth and balanced robot-environment interaction with proper force directions. Finally, the proposed approach provides adequate force feedback about the environment to the operator and assists the operator in identifying different motion situations of the slaves. Our approach demonstrates that the forces from the slaves will not interrupt the operator's perception of the environment. To validate the proposed system, experiments are conducted using a platform consisting of two 7-DoF slave robots and one 3-DoF master haptic device. The experiments demonstrated good results in terms of position, orientation, and force regulation.

In this paper, we present a model to propose an optimal placement for a robot in a social group interaction. Our model estimates the O-space according to the F-formation theory. The method automatically calculates a suitable placement for the robot. An evaluation of the method has been performed by conducting an experiment where participants stand in different formations and a robot is teleoperated to join the group. In one condition, the operator positions the robot according to the specified location given by our algorithm. In another condition, operators have the freedom to position the robot according to their personal choice. Follow-up questionnaires were performed to determine which of the placements were preferred by the participants. The results indicate that the proposed method for automatic placement of the robot is supported from the participants. The contribution of this work resides in a novel method to automatically estimate the best placement of the robot, as well as the results from user experiments to verify the quality of this method. These results suggest that teleoperated robots such as mobile robot telepresence systems could benefit from tools that assist operators in placing the robot in groups in a socially accepted manner.

This article presents an approach for assisted teleoperation of a robot arm, formulated within a real-time stack-of-tasks (SoT)whole-body motion control framework. The approach leverages the hierarchical nature of the SoT framework to integrate operator commands with assistive tasks, such as joint limit and obstacle avoidance or automatic gripper alignment. Thereby some aspects of the teleoperation problem are delegated to the controller and carried out autonomously. The key contributions of this work are two-fold: the first is a method for unobtrusive integration of autonomy in a telemanip-ulation system; and the second is a user study evaluation of the proposed system in the context of teleoperated pick-and-place tasks. The proposed approach of assistive control was found to result in higher grasp success rates and shorter trajectories than achieved through manual control, without incurring additional cognitive load to the operator.

There is an increasing need to derive semantics from real-world observations to facilitate natural information sharing between machine and human. Conceptual spaces theory is a possible approach and has been proposed as mid-level representation between symbolic and sub-symbolic representations, whereby concepts are represented in a geometrical space that is characterised by a number of quality dimensions. Currently, much of the work has demonstrated how conceptual spaces are created in a knowledge-driven manner, relying on prior knowledge to form concepts and identify quality dimensions. This paper presents a method to create semantic representations using data-driven conceptual spaces which are then used to derive linguistic descriptions of numerical data. Our contribution is a principled approach to automatically construct a conceptual space from a set of known observations wherein the quality dimensions and domains are not known a priori. This novelty of the approach is the ability to select and group semantic features to discriminate between concepts in a data-driven manner while preserving the semantic interpretation that is needed to infer linguistic descriptions for interaction with humans. Two data sets representing leaf images and time series signals are used to evaluate the method. An empirical evaluation for each case study assesses how well linguistic descriptions generated from the conceptual spaces identify unknown observations. Furthermore, comparisons are made with descriptions derived on alternative approaches for generating semantic models.

This paper depicts a sensor-based map navigation approach which targets users, who due to disabilities or lack of technical knowledge are currently not in the focus of map system developments for personalized information. What differentiates our approach from the state-of-art mostly integrating localized social media data, is that our vision is to integrate real time sensor generated data that indicates the situation of different phenomena (such as the physiological functions of the body) related to the user. The challenge hereby is mainly related to knowledge representation and integration. The tentative impact of our vision for future navigation systems is reflected within a scenario.

There is an increasing interest in the machine learning community to automatically learn feature representations directly from the (unlabeled) data instead of using hand-designed features. The autoencoder is one method that can be used for this purpose. However, for data sets with a high degree of noise, a large amount of the representational capacity in the autoencoder is used to minimize the reconstruction error for these noisy inputs. This paper proposes a method that improves the feature learning process by focusing on the task relevant information in the data. This selective attention is achieved by weighting the reconstruction error and reducing the influence of noisy inputs during the learning process. The proposed model is trained on a number of publicly available image data sets and the test error rate is compared to a standard sparse autoencoder and other methods, such as the denoising autoencoder and contractive autoencoder.

Mining and representation of qualitative patterns is a growing field in sensor data analytics. This paper leverages from rule mining techniques to extract and represent temporal relation of prototypical patterns in clinical data streams. The approach is fully data-driven, where the temporal rules are mined from physiological time series such as heart rate, respiration rate, and blood pressure. To validate the rules, a novel similarity method is introduced, that compares the similarity between rule sets. An additional aspect of the proposed approach has been to utilize natural language generation techniques to represent the temporal relations between patterns. In this study, the sensor data in the MIMIC online database was used for evaluation, in which the mined temporal rules as they relate to various clinical conditions (respiratory failure, angina, sepsis, …) were made explicit as a textual representation. Furthermore, it was shown that the extracted rule set for any particular clinical condition was distinct from other clinical conditions.

One goal of mobile robotic telepresence for social interaction is to design robotic units that are easy to operate for novice users and promote good interaction between people. This paper presents an exploratory study on the effect of camera orientation and field of view on the interaction between a remote and local user. Our findings suggest that limiting the width of the field of view can lead to better interaction quality as it encourages remote users to orient the robot towards local users.

Mobile robotic telepresence (MRP) has been introduced to allow communication from remote locations. Modern MRP systems offer rich capabilities for human-human interactions. However, simply driving a telepresence robot can become a burden especially for novice users, leaving no room for interaction at all. In this video we introduce a project which aims to incorporate advanced robotic algorithms into manned telepresence robots in a natural way to allow human-robot cooperation for safe driving. It also shows a very first implementation of cooperative driving based on extracting a safe drivable area in real time using the image stream received from the robot.

This position paper introduces the utilityof the conceptual spaces theory to conceptualisethe acquired knowledge in data-totextsystems. A use case of the proposedmethod is presented for text generationsystems dealing with sensor data. Modellinginformation in a conceptual spaceexploits a spatial representation of domainknowledge in order to perceive unexpectedobservations. This ongoing work aimsto apply conceptual spaces in NLG forgrounding numeric information into thesymbolic representation and confrontingthe important step of acquiring adequateknowledge in data-to-text systems.

Early detection and adaptive support to changing individual needs related to ageing is an important challenge in today's society. In this paper we present a system called GiraffPlus that aims at addressing such a challenge and is developed in an on-going European project. The system consists of a network of home sensors that can be automatically configured to collect data for a range of monitoring services; a semi-autonomous telepresence robot; a sophisticated context recognition system that can give high-level and long term interpretations of the collected data and respond to certain events; and personalized services delivered through adaptive user interfaces for primary users. The system performs a range of services including data collection and analysis of long term trends in behaviors and physiological parameters (e.g. relating to sleep or daily activity); warnings, alarms and reminders; and social interaction through the telepresence robot. The latter is based on the Giraff telepresence robot, which is already in place in a number of homes. A distinctive aspect of the project is that the GiraffPlus system will be installed and evaluated in at least 15 homes of elderly people. This paper provides a general overview of the GiraffPlus system and its evaluation.

Fast growing structured knowledge in machine processable formats such as RDF/OWL provides the opportunity of having automatic annotation for stream data in order to extract meaningful information. In this work, we propose a system architecture to model the process of stream data annotation in an automatized fashion using public repositories of knowledge. We employ abductive reasoning which is capable of retrieving the best explanations for observations given incomplete knowledge. In order to evaluate the effectiveness of the framework, we use multivariate data coming from medical sensors observing a patient in ICU (Intensive Care Unit) suffering from several diseases as the ground truth against which the eventual explanations (annotations) of the reasoner are compared.

Health monitoring systems using wearable sensors have rapidly grown in the biomedical community. The main challenges in physiological data monitoring are to analyse large volumes of health measurements and to represent the acquired information. Natural language generation is an effective method to create summaries for both clinicians and patients as it can describe useful information extracted from sensor data in textual format. This paper presents a framework of a natural language generation system that provides a text-based representation of the extracted numeric information from physiological sensor signals. More specifically, a new partial trend detection algorithm is introduced to capture the particular changes and events of health parameters. The extracted information is then represented considering linguistic characterisation of numeric features. Experimental analysis was performed using a wearable sensor and demonstrates a possible output in natural language text.

Perceptual anchoring deals with the problem of creating and maintaining the connection between percepts and symbols that refer to the same physical object. When approaching long term use of an anchoring framework which must cope with large sets of data, it is challenging to both efficiently and accurately anchor objects. An approach to address this problem is through visual perception and computationally efficient binary visual features. In this paper, we present a novel hash table algorithm derived from summarized binary visual features. This algorithm is later contextualized in an anchoring framework. Advantages of the internal structure of proposed hash tables are presented, as well as improvements through the use of hierarchies structured by semantic knowledge. Through evaluation on a larger set of data, we show that our approach is appropriate for efficient bottom-up anchoring, and performance-wise comparable to recently presented search tree algorithm.

Robotic telepresence, also known as telerobotics is a subfield of telepresence whose aim is to increase presence via embodiment in a robotic platform. In particular, robotic telepresence can be an effective tool to enhance social interaction suited to certain groups of users such as the elderly. The aim of this workshop is to address various aspects important for social robotic telepresence which include but are not limited to, (1) the mechanical design, (2) the user interface design, (3) the interaction between the remotely embodied person and the locally embodied person and (4) the perception of social robotic telepresence systems. Furthermore, we are interested in discovering the added value of spatial presence in the context of social telepresence and comparisons between robotic and non-robotic systems are of interest. We welcome contributions concerning results reached from the above mentioned areas of interest, user evaluation and methodologies, as well as reports from the deployment of social robotic solutions into real world contexts.

In this paper we analyze the problem of discrimination of gases with mobile robots. Previously, it has been shown that the conditions in which data is collected heavily influence the characteristics of the signal to be identified. As a result, the already difficult task of selecting features which characterize a gas is made more challenging by the absence of a steady state response. This is often due to the movement of the robot, and/or the physical properties of the environment, e.g., turbulent airflow creating patches and eddies in the plume. In this work we compare two approaches for feature selection which are able to consider explicitly the information on the experimental setup and optimize the subset of features used in the recognition process. The approaches are tested on a large data set collected with a mobile robot moving in different environments (outdoors and indoors). The results show that the classification performance is improved resulting in a higher average accuracy and lower variance in the accuracy across the different experimental setups.

In this paper we describe an implemented framework that integrates knowledge representation and reasoning in a symbiotic system. In such systems a number of heterogeneous sensors pervasively embedded in the environment, mobile robots and humans co-exist and communicate. In this work, the integration is mediated through perceptual anchoring, which creates and maintains the correspondences between the symbol system and the perceptual data that refer to the same physical object. The overall framework is evaluated using ResearchCyc as the knowledge representation and reasoning system, within the context of a physical testbed, which consists of a small apartment-like home.

In this paper we investigate how a mobile robot equipped with tin dioxide gas sensors and an anemometer can use an online classification algorithm in order to improve the exploration strategy. The purpose of the platform is to establish the character of a gas source with accuracy while minimizing the time required for exploration. For this to be possible, the output of the classification algorithm is probabilistic, feeding in a sequence of posterior probabilities to a path planner. To further assist path planning, a 3d-ultrasonic anemometer is available which give indication on the average wind speed and direction. In addition to evaluating different olfaction driven path planning strategies, experimental validations also evaluate the classification algorithms and its application to different environments with varying characteristics.

Classification of odours with an array of gas sensors mounted on a mobile robot is a challenging and still relatively unexplored topic. Mobile robots able to classify an odour could navigate to a specific source or isolate high concentration areas in applications such as environmental monitoring. A key aspect to classification is to be able to process the data collected while moving the robot and using a simple and compact sensor system. In order to achieve this, we present a classification algorithm that is based on the transient response from the sensors. An analysis of how classification results vary with regards to the movement of the robot is provided and subsequently the experimental validations show that the classification performance depends more on how the robot traverses the odour plume and the quality of the transient than on the distance from the source location. The experimental validation has been done in a large unmodified indoor environment.

In this paper we present initial experiments towards environmental monitoring with a mobile platform. A prototype of a pollution monitoring robot was set up which measures the gas distribution using an ldquoelectronic noserdquo and provides three dimensional wind measurements using an ultrasonic anemometer. We describe the design of the robot and the experimental setup used to run trials under varying environmental conditions. We then present the results of the gas distribution mapping. The trials which were carried out in three uncontrolled environments with very different properties: an enclosed indoor area, a part of a long corridor with open ends and a high ceiling, and an outdoor scenario are presented and discussed.

In this work we report results on the use of symbolic knowledge representation and reasoning (KRR) for perceptual anchoring. Anchoring is the creation and maintenance of a connection between the symbolic and perceptual description that refer to the same physical object in the environment. We extend the anchoring framework to manage the symbolic information in a KRR system, and to exploit this knowledge and the inference mechanism to recover from failures in the anchoring of symbols. We show a simulated scenario where the system communicates with a user to interactively resolve an ambiguous description using the knowledge base, and in particular spatial relations. This is a first step towards a KRR-supported anchoring framework that we will use for human-robot communication.

Robot ecologies provide a new paradigm for assistive, service, industrial, and entertainment robotics which is quickly gaining popularity. These ecologies contain a large number of robotic components pervasively embedded in the environment and interacting with each other. Human users of such systems need to be able to interface with both the system as a w hole and, if desired, which each individual component. The humans should be able to transmit, in a natural way, commands that range from basic ones, such as "turn on the lights in the bedroom", to abstract ones, such as "bring me a cup of coffee". Human users may also need to interact with task execution, especially at decision points. In this paper, we introduce an approach to interface a human user to a specific type of robot ecology, called an ecology of Physically Embedded Intelligent Systems, or PEIS-Ecology. The ecology includes simple sensors and actuators and more complicated devices such as mobile robots. The proposed interface satisfies two requirements: 1) to easily and automatically generate component interfaces, and 2) to provide a simple mechanism by which to request and monitor the execution of tasks in the ecology.

We present a new approach for odour detection and recognition based on a so-called PEIS-Ecology: a network of gas sensors and a mobile robot are integrated in an intelligent environment. The environment can provide information regarding the location of potential odour sources, which is then relayed to a mobile robot equipped with an electronic nose. The robot can then perform a more thorough analysis of the odour character. This is a novel approach which alleviates some the challenges in mobile olfaction techniques by single and embedded mobile robots. The environment also provides contextual information which can be used to constrain the learning of odours, which is shown to improve classification performance

More and more work in the field of artificial olfaction considers the integration of olfaction onto robotic systems. An important part of this integration is providing the robot with the ability to discriminate between different odour substances. This work presents the integration of an electronic nose onto a complete robotic system with multi-sensing modalities. The ambition of this work is to illustrate how classification performance of odours can be improved through the exploitation of the mobility of a robot, as well as through the cooperation between a human user and an electronic perceptual system. These points are highlighted in the context of a online robotic system designed to perform different tasks which require the ability to discriminate between odour characters. The robotic system and experimental results are presented where these tasks are tested and evaluated.

Olfaction is a challenging new sensing modality for intelligent systems. With the emergence of electronic noses it is now possible to detect and recognise a range of different odours for a variety of applications. An existing application is to use electronic olfaction on mobile robots for the purpose of odour based navigation. In this work, we introduce a new application where electronic olfaction is used in cooperation with other types of sensors on a mobile robot in order to acquire the odour property of objects. The mobility of the robot facilitates the execution of specific perceptual actions, such as moving closer to objects to acquire odour properties. Additional sensing modalities provides the spatial detection of objects and electronic olfaction then acquires the odour property which can be used for discrimination and recognition of the object being considered. We examine the problem of deciding when, how and where the e-nose should be activated by planning for active perception. We investigate the use of symbolic reasoning techniques in this context and consider the problem of integrating the information provided by the e-nose with both prior information and information from other sensors (e.g., vision). Finally, experiments are performed on a mobile robot equipped with an e-nose together with a variety of sensors that can perform decision making tasks in realistic environments.