We present a CNN-based approach for scene-level change detection in aerial images with registration errors. Thousands of aerial images and long videos are routinely acquired for monitoring large areas such as forests and oil pipelines. Annotating changes in those videos and images can be tedious, error-prone, or even unfeasible for a human operator. Moreover, accurate pixel-wise registration is usually unavailable, and conventional descriptor-based registration methods are doomed to fail since they rely on similarities to establish correspondences that are impaired due to the latent changes in the scene. We introduce a new neural network architecture that can be trained end-to-end to simultaneously perform image registration and change detection to mitigate these issues. Our approach reduces the number of parameters required to optimize the process while steeping towards a more robust change detection pipeline for UAV images. We evaluated our method in two datasets containing registration errors of up to 96 pixels in translation and 30° in rotation. The results showed that our approach outperformed the state-of-the-art, achieving a 6% improvement in the AUC metric.

Despite the advances in autonomous navigation and motion planning, there are still several challenges to overcome, especially for confined or underground spaces. Confined scenarios present challenges such as lack of global or accurate external localization, uneven and slippery terrains, and multilevel stages. Exploring and mapping unknown unstructured environments is a fundamental step into the safe and efficient accomplishment of real-world tasks such as search and rescue missions or the autonomous inspection of dangerous areas. This paper proposes a novel three-dimensional autonomous exploration method for ground robots that considers the terrain traversability combined with the frontier expected information gain as a metric for the next best frontier selection in GPS-denied, confined spaces. Safe paths for navigation and frontier extraction are calculated iteratively from multiple 3D map representations such as octrees and meshes. Results in realistic simulated underground scenarios from the DARPA subterranean challenge demonstrate the technique's feasibility, achieving a more reliable and faster exploration rate over competing approaches.

Technological advances in sensors have paved the way for digital cameras to become increasingly ubiquitous, which, in turn, led to the popularity of the self-recording culture. As a result, the amount of visual data on the Internet is moving in the opposite direction of the available time and patience of the users. Thus, most of the uploaded videos are doomed to be forgotten and unwatched stashed away in some computer folder or website. In this paper, we address the problem of creating smooth fast-forward videos without losing the relevant content. We present a new adaptive frame selection formulated as a weighted minimum reconstruction problem. Using a smoothing frame transition and filling visual gaps between segments, our approach accelerates first-person videos emphasizing the relevant segments and avoids visual discontinuities. Experiments conducted on controlled videos and also on an unconstrained dataset of First-Person Videos (FPVs) show that, when creating fast-forward videos, our method is able to retain as much relevant information and smoothness as the state-of-the-art techniques, but in less processing time.

Different applications in remote sensing, such as crop monitoring and visual surveillance, demand the automatic detection of changes from sets of images acquired over time. Most traditional approaches use satellite imagery, which, besides the known issues such as cloud cover and image acquisition frequency for nongeostationary satellites, are very costly. In this context, with the recent technological advances, unmanned aerial vehicles (UAVs) have become ubiquitous in numerous applications. In this letter, we present a fully convolutional Siamese autoencoder method for change detection in aerial images, in particular for those obtained with UAVs. We show that, by using an autoencoder, we can further reduce the number of labeled samples required to achieve competitive results. We evaluated the performance of our approach on two different data sets, and the results showed that our methodology outperforms the state of the art, while demanding less training data.

At the core of most three-dimensional alignment and tracking tasks resides the critical problem of point correspondence. In this context, the design of descriptors that efficiently and uniquely identifies keypoints, to be matched, is of central importance. Numerous descriptors have been developed for dealing with affine/perspective warps, but few can also handle non-rigid deformations. In this paper, we introduce a novel binary RGB-D descriptor invariant to isometric deformations. Our method uses geodesic isocurves on smooth textured manifolds. It combines appearance and geometric information from RGB-D images to tackle non-rigid transformations. We used our descriptor to track multiple textured depth maps and demonstrate that it produces reliable feature descriptors even in the presence of strong non-rigid deformations and depth noise. The experiments show that our descriptor outperforms different state-of-the-art descriptors in both precision-recall and recognition rate metrics. We also provide to the community a new dataset composed of annotated RGB-D images of different objects (shirts, cloths, paintings, bags), subjected to strong non-rigid deformations, to evaluate point correspondence algorithms.

In this paper, we introduce a novel approach for semantic description of object features based on the prototypicality effects of the Prototype Theory. Our prototype-based description model encodes and stores the semantic meaning of an object, while describing its features using the semantic prototype computed by CNN-classifications models. Our method uses semantic prototypes to create discriminative descriptor signatures that describe an object highlighting its most distinctive features within the category. Our experiments show that: i) our descriptor preserves the semantic information used by the CNN-models in classification tasks; ii) our distance metric can be used as the object's typicality score; iii) our descriptor signatures are semantically interpretable and enables the simulation of the prototypical organization of objects within a category.

Thanks to the advances in the technology of low-cost digital cameras and the popularity of the self-recording culture, the amount of visual data on the Internet is going to the opposite side of the available time and patience of the users. Thus, most of the uploaded videos are doomed to be forgotten and unwatched in a computer folder or website. In this work, we address the problem of creating smooth fast-forward videos without losing the relevant content. We present a new adaptive frame selection formulated as a weighted minimum reconstruction problem, which combined with a smoothing frame transition method accelerates first-person videos emphasizing the relevant segments and avoids visual discontinuities. The experiments show that our method is able to fast-forward videos to retain as much relevant information and smoothness as the state-of-the-art techniques in less time. We also present a new 80-hour multimodal (RGB-D, IMU, and GPS) dataset of first-person videos with annotations for recorder profile, frame scene, activities, interaction, and attention.

In this work, multiple robots circulate around the boundary of a desired region in order to create a virtual fence. The aim of the this fence is to avoid internal or external agents crossing through the delimited area. In this paper, we propose a distributed technique that allows a team of robots to plan the deformation of the boundary shape in order to escort the safe region from one place to a goal. Our proposal is composed of two parts. First, we present a distributed planning method for the dynamic boundary. We model the resulting plan as a twice differentiable function. Second, we use the obtained function to guide the robot team, where every member uses only local information for the controller. The robots distribute themselves along the time-varying perimeter and patrol around it. We show in simulation how the robots behave in partially/totally unknown environments with static obstacles.

Recent work in the field of bio-inspired robotic systems has introduced designs for modular robots that are able to assemble into structures (e.g., bridges, landing platforms, fences) using their bodies as the building components. Yet, it remains an open question as to how to program large swarms of robotic modules so that the assembly task is performed as efficiently as possible. Moreover, the problem of designing assembly algorithms is compounded by the scale of these systems, and by the lack of centralized guidance in unstructured environments. The main contribution of this work is a decentralized algorithm to assemble structures with modular robots. Importantly, we coordinate the robots so that docking actions can be parallelized. We show the correctness of our algorithm, and we demonstrate its scalability and generality through multiple scenarios in simulation. Experiments on physical robots demonstrate the validity of our approach in real-world settings.

Thanks to the low operational cost and large storage capacity of smartphones and wearable devices, people are recording many hours of daily activities, sport actions and home videos. These videos, also known as egocentric videos, are generally long-running streams with unedited content, which make them boring and visually unpalatable, bringing up the challenge to make egocentric videos more appealing. In this work we propose a novel methodology to compose the new fast-forward video by selecting frames based on semantic information extracted from images. The experiments show that our approach outperforms the state-of-the-art as far as semantic information is concerned and that it is also able to produce videos that are more pleasant to be watched.

In this paper, we propose a motion planning method to escort a set of agents from one place to a goal in an environment with obstacles. The agents are distributed in a finite area, with a time-varying perimeter, in which we put multiple robots to patrol around it with a desired velocity. Our proposal is composed of two parts. The first one generates a plan to move and deform the perimeter smoothly, and as a result, we obtain a twice differentiable boundary function. The second part uses the boundary function to compute a trajectory for each robot, we obtain each resultant trajectory by first solving a differential equation. After receiving the boundary function, the robots do not need to communicate among themselves until they finish their trajectories. We validate our proposal with simulations and experiments with actual robots.

In this paper we propose a new vision-based obstacle avoidance strategy using the Underwater Dark Channel Prior (UDCP) that can be applied to any Unmanned Underwater Vehicle (UUV) equipped with a simple monocular camera and minimal on-board processing capabilities. For each incoming image, our method first computes a relative depth map to estimate the obstacles nearby. Then, the map is segmented and the most promising Region of Interest (RoI) is identified. Finally, an escape direction is computed within the RoI and a control action is performed accordingly to avoid the obstacles. We tested our approach on a video sequence in a natural environment and compared it against a state-of-the-art method showing better performance, specially in light changing conditions. We also provide online results on a low-cost Remotely Operated Vehicle (ROV) in a controlled environment.

This paper addresses a method for stabilizing the attitude of a quadrotor-like Unmanned Aerial Vehicle (UAV). Towards this end, a new fuzzy control technique based on Linear Matrix Inequalities (LMIs) is proposed, relying on a simple Takagi-Sugeno (TS) fuzzy model when compared to others found in the Literature. This strategy guarantees prescribed performance criteria, such as exponential decay rate, allowing stability under aggressive disturbances. The designed local gains are combined through membership functions of the model and are used to regulate the attitude of the vehicle on the hovering operating point, an important flight mode for this kind of robot. Numerical analysis reveal how the transient response can be tuned and tests with disturbances show the stability of the vehicle starting from different initial conditions, illustrating the merits of the proposed approach.

This paper presents a method for stabilizing the attitude of a Hybrid Unmanned Aerial Underwater Vehicle. Firstly, we present aerodynamic and hydrodynamic models for the angular motion of our robot, discussing effects like buoyancy force and added inertia. Next, we apply robust control techniques for both environment, aerial and underwater, based on linear uncertain models with only four vertices and well-defined stability criteria, such as D-stability and ℋ2performance. Gain matrices Kairand Kwatare computed and the attitude of the vehicle at the hovering operation point for each environment is controlled, respectively. Finally, a procedure is proposed to check the global stability for the switching control case, when the robot changes from air to water (or vice-versa). Numerical simulations with disturbances and switching control are presented to show the stability at different initial conditions.

In many cases, large area disasters could be possibly be prevented if the incipient small-scale anomalies are detected in their early stages. A way to accomplish this would be to have multiple sensors deployed in disaster prone areas to detect anomalies. However, compared to static sensor networks, robotic sensor networks offer advantages such as active sensing, large area coverage and anomaly tracking. This paper addresses the problem of coordinating and controlling multiple robots for the detection of multiple dynamic anomalies in the environment. The main contribution of the work is a combined approach for the effective exploration under uncertainty, the anomaly tracking, and the autonomous on-line allocation of agents. Robots explore the work area maintaining the history of the sensed areas to reduce redundancy and to allow for full-map coverage. When an anomaly is detected, a robot autonomously determines how to either track the anomaly or to continue the exploration of the environment, depending on the size of the anomaly, which is estimated by the length of the perimeter of the enclosing polygon. We show results of our methodology both in simulation and with actual robots which have demonstrated that robots can autonomously and distributively be allocated to track or to explore depending on the behavior of the detected anomalies.

Underwater environments present a considerable challenge for computer vision, since water is a scattering medium with substantial light absorption characteristics which is made even more severe by turbidity. This poses significant problems for visual underwater navigation, object detection, tracking and recognition. Previous works tackle the problem by using unreliable priors or expensive and complex devices. This paper adopts a physical underwater light attenuation model which is used to enhance the quality of images and enable the applicability of traditional computer vision techniques images acquired from underwater scenes. The proposed method simultaneously estimates the attenuation parameter of the medium and the depth map of the scene to compute the image irradiance thus reducing the effect of the medium in the images. Our approach is based on a novel optical flow method, which is capable of dealing with scattering media, and a new technique that robustly estimates the medium parameters. Combined with structure-from-motion techniques, the depth map is estimated and a model-based restoration is performed. The method was tested both with simulated and real sequences of images. The experimental images were acquired with a camera mounted on a Remotely Operated Vehicle (ROV) navigating in a naturally lit, shallow seawater. The results show that the proposed technique allows for substantial restoration of the images, thereby improving the ability to identify and match features, which in turn is an essential step for other computer vision algorithms such as object detection and tracking, and autonomous navigation.

Path planning is a fundamental task for any kind of autonomous mobile robot. In this paper, we present a motion planning technique on the three-dimensional space considering vehicles with spatial curvature and pitch (climb or dive) angle constraints. Concerning real fixed-wing Unmanned Aerial Vehicles or underwater Remotely Operated Vehicles, we face the problem of calculating paths with continuous acceleration profiles, preventing the robot from abrupt turning or climbing/diving rates during its navigation. We also focus on the generation of paths with reduced length, allowing energy savings and minimizing the time for the task fulfillments. The proposed methodology provides, in a fast and iterative way, near-optimal paths in some scenarios, but with the advantage of continuous curvature and pitch rate profiles, a fundamental issue regarding the dynamics of real-world robots. The methodology was validated through numerous trials, under different scenarios in a simulated environment, and the results compared to the ones obtained by state-of-the-art techniques, providing a thorough evaluation.

This paper proposes a novel methodology to estimate the optical flow in scattering media, which consists on new formulation based on the classical Horn-Schunk approach and the optical image formation model. Our formulation is able to deal with the hard problem of tracking points in a medium where there is absorption and scattering effects. This approach generalizes assumptions of the Horn-Schunk model in order to tackle both non-scattering and scattering media. Our approach uses the Dark Channel Prior to estimate the scene transmission, which attains a significant improvement in the optical flow estimation in scattering media. We show that our approach outperformed state-of-the-art models and we provide a detailed analysis of our technique that shows its applicability to image sequences acquired both in simulated and real scenes.

The complete modeling and simulation of an unmanned vehicle with combined aerial and underwater capabilities, called Hybrid Unmanned Aerial Underwater Vehicle (HUAUV), is presented in this paper. The best architecture for this kind of vehicle was evaluated based on the adaptation of typical platforms for aerial and underwater vehicles, to allow the navigation in both environments. The model selected was based on a quadrotor-like aerial platform, adapted to dive and move underwater. Kinematic and dynamic models are presented here, and the parameters for a small dimension prototype was estimated and simulated. Finally, controllers were used and validated in realistic simulation, including air and water navigation, and the environment transition problem. To the best of our knowledge, it is the first vehicle that is able to navigate in both environment without mechanical adaptation during the medium transitions.

The ability to detect changes is an essential competence that robots should possess for increased autonomy. In several applications, such as surveillance, a robot needs to detect relevant changes in the environment by comparing current sensory data with previously acquired information from the environment. We present an efficient method for point cloud comparison and change detection in 3D environments based on spatial density patterns. Our method automatically segments 3D data corrupted by noise and outliers into an implicit volume bounded by a surface, making it possible to efficiently apply Boolean operations in order to detect changes and to update existing maps. The method has been validated on several trials using mobile robots operating in real environments and its performance was compared to state-of-the-art algorithms. Our results demonstrate the performance of the proposed method, both in greater accuracy and reduced computational cost.

Successful image-based object recognition techniques have been constructed founded on powerful techniques such as sparse representation, in lieu of the popular vector quantization approach. However, one serious drawback of sparse space-based methods is that local features that are quite similar can be quantized into quite distinct visual words. We address this problem with a novel approach for object recognition, called sparse spatial coding, which efficiently combines a sparse coding dictionary learning and spatial constraint coding stage. We performed experimental evaluation using the Caltech 101, Caltech 256, Corel 5000, and Corel 10000 data sets, which were specifically designed for object recognition evaluation. Our results show that our approach achieves high accuracy comparable with the best single feature method previously published on those databases. Our method outperformed, for the same bases, several multiple feature methods, and provided equivalent, and in few cases, slightly less accurate results than other techniques specifically designed to that end. Finally, we report state-of-the-art results for scene recognition on COsy Localization Dataset (COLD) and high performance results on the MIT-67 indoor scene recognition, thus demonstrating the generalization of our approach for such tasks.

In applications such as surveillance, pedestrian detection can be seen as a filtering stage which will locate the objects of interest so that higher level tasks, such as recognition, re-identification, action and activity recognition, can be performed considering only those objects. Therefore, it is imperative that the pedestrian detection task presents low computational cost. Several methods have been proposed to detect pedestrians in images and videos. However, a remaining challenge is to detect pedestrians with high accuracy at a very low computational cost. Towards accomplishing the goal of reducing the costs for pedestrian detection, we propose a cascade of rejection based on Partial Least Squares (PLS) and the variable selection method Variable Importance in Projection (VIP) combined with the propagation of latent variables through the stages. Our results show that the method reduces the computational cost by increasing the number of rejected background samples in earlier stages of the cascade.

In this paper, we introduce the k-Dubins Traveling Salesman Problem with Neighborhoods (k-DTSPN), the problem of planning efficient paths among target regions for multiple robots with bounded curvature constraints (Dubins vehicles). This paper presents two approaches for the problem. Firstly, we present a heuristic that solves it in two steps, based on classical techniques found in the literature. Secondly, we employ a Memetic Algorithm to solve both combinatorial and continuous phases of the problem in a combined manner. We provide formal analysis about both proposed techniques, presenting upper bounds to the length of the longest tour. Numerous trials in simulated environments were executed, providing statistical examination of the final results.

A Wireless Sensor Network consists of several sensor nodes deployed in an environment having as primary goal to collect data. However, due to limited sensor communication range, oftentimes it is necessary to use a mobile node that will visit other nodes to gather up their collected data. This work addresses the problem of planning efficient paths for data collection by a mobile node modeled as a nonholonomic vehicle with curvature constraints. We propose an efficient algorithm to identify areas of intersection among the nodes RF footprints which will guide the identification of a smaller set of waypoints through which the vehicle needs to traverse in order to collect available data. Then, a metric similar to the classical Traveling Salesman Problem is used to determine the best circuit that includes all these collecting points. In order to reduce the total path length for the mobile node, a meta-heuristic is used. The classical Dubins' path technique is employed to generate a feasible tour for the vehicle and a new heuristic is used to generate the required orientation at the collecting points. The methodology was validated in a simulated environment. Our methodology outperforms the classical Alternating Algorithm and the best performing state-of-the-art algorithm.

In this work we propose an efficient and simple three-stage evolutionary algorithm to tackle the difficult problem of planning shorter paths through regions of an environment which are feasible for a nonholonomic vehicle with curvature constraints (e.g. Dubins' vehicle). Our method is able to efficiently solve both the combinatorial and the continuous steps of the problem in a combined manner. In the first phase, the method varies the position of the waypoints within the boundaries of each region, it then optimizes the path orientation at each waypoint, and finally it chooses the best actual sequence of visit. Numerous trials, under different scenarios in a simulated environment, were executed providing a thorough evaluation and validation of the methodology. The results show that a substantial improvement was obtained on the search for optimal paths in the DTSPN over current works in the literature. Numerical simulations also exhibit a significant performance improvement when compared with classical solutions that use the Alternating Algorithm, and they also show that our method outperforms a random sampling based technique. Our results present a reduction on the final path length of about 25% on average when compared to paths generated by the aforementioned methods.

Successful state-of-the-art object recognition techniques from images have been based on powerful methods, such as sparse representation, in order to replace the also popular vector quantization (VQ) approach. Recently, sparse coding, which is characterized by representing a signal in a sparse space, has raised the bar on several object recognition benchmarks. However, one serious drawback of sparse space based methods is that similar local features can be quantized into different visual words. We present in this paper a new method, called Sparse Spatial Coding (SSC), which combines a sparse coding dictionary learning, a spatial constraint coding stage and an online classification method to improve object recognition. An efficient new off-line classification algorithm is also presented. We overcome the problem of techniques which make use of sparse representation alone by generating the final representation with SSC and max pooling, presented for an online learning classifier. Experimental results obtained on the Caltech 101, Caltech 256, Corel 5000 and Corel 10000 databases, show that, to the best of our knowledge, our approach supersedes in accuracy the best published results to date on the same databases. As an extension, we also show high performance results on the MIT-67 indoor scene recognition dataset.

The ability to detect changes in the environment is an essential trait for robots commissioned to work in several applications. In surveillance, for instance, a robot needs to detect meaningful changes in the environment which is achieved by comparing current sensory data with previously acquired information from the environment. The large amount of sensory data, which are often complex and very noisy, explains the inherent difficulty of this task. As an attempt to tackle this hard problem, we present an efficient method to automatically segment 3D data, corrupted with noise and outliers, into an implicit volume bounded by a surface. The method makes it possible to efficiently apply Boolean operations to 3D data in order to detect changes and to update existing maps. We show that our approach is powerful, albeit simple, with linear time complexity. The method has been validated through several trials using mobile robots operating in real environments and their performance was compared to another state-of-art algorithm. Experimental results demonstrate the performance of the proposed method, both in accuracy and computational cost.

This work introduces a novel descriptor called Binary Robust Appearance and Normals Descriptor (BRAND), that efficiently combines appearance and geometric shape information from RGB-D images, and is largely invariant to rotation and scale transform. The proposed approach encodes point information as a binary string providing a descriptor that is suitable for applications that demand speed performance and low memory consumption. Results of several experiments demonstrate that as far as precision and robustness are concerned, BRAND achieves improved results when compared to state of the art descriptors based on texture, geometry and combination of both information. We also demonstrate that our descriptor is robust and provides reliable results in a registration task even when a sparsely textured and poorly illuminated scene is used.

In this paper we present an improved color-based planar fiducial marker system. Our framework provides precise and robust full 3D pose estimation of markers with superior accuracy when compared with many fiducial systems in the literature, while color information encoding enables using over 65 000 distinct markers. Unlike most color-based fiducial frameworks, which requires prior classification training and color calibration, ours can perform reliably under illumination changes, requiring but a rough white balance adjustment. Our methodology provides good detection performance even under poor illumination conditions which typically compromise other marker identification techniques, thus avoiding the evaluation of otherwise falsely identified markers. Several experiments are presented and carefully analyzed, in order to validate our system and demonstrate the significant improvement in estimation accuracy of both position and orientation over traditional techniques.

In this paper we deal with the problem of path length optimization for nonholonomic robots modeled as Dubins' vehicles. We present an improved solution that computes paths through two-dimensional waypoints, that are shorter than those produced by one of the most used techniques in the literature. We initially present an optimization cost function that proves to be better suited for the nonholonomic constraints of the vehicles that are the focus of this work. We also propose an improvement for the Alternating Algorithm, used to determinate the orientation angles on the calculation of the Dubins' Path through the waypoints. Finally, we use a recently developed optimization meta-heuristic, called Continuous-GRASP (C-GRASP), to generate a path that is shorter than paths obtained with classical techniques. Our results show significant improvements on the search for optimal paths for the case of nonholonomic vehicles. Our methodology was thoroughly evaluated and validated in simulation, and the results have shown a decrease on path's length of 28% on average, compared with the classic technique found in literature. In some cases a reduction of approximately 50% was obtained.

Dance is a way of artistic expression. It also can be considered a form of communication. Typically, dance relies on the capability that human beings have to synchronize their movements rhythmically to music. In this work, we propose a music-driven approach to enable mobile robots to dance to a given music. We focus on the automatic generation of paths such that robots jointly performs a choreography guided by a song. Although dancing ability of a real person is complex and hard to accomplish, in this paper we aim to associate the robot motion to music and to emulate this typical human activity. We propose a variability function for evaluating the relevance of a block (set of notes) in a particular song, and based on those results, similar blocks are clustered. Finally, we generate customized paths for robots considering the current sound (note) being played and the importance of the block to which this note belongs. We have tested this approach with real robots (e-puck) using different kinds of songs. Results from those experiments indicate that robots can achieve good choreographic performance even if dance steps are automatically generated without previous knowledge of the song style.

Localizing a user is a fundamental problem that arises in many potential applications. The use of wireless technologies for locating a user has been a trend in recent years. Most existing approaches use RSSI to localize the user. In general, one of the several existing wireless standards such as ZigBee, Bluetooth or Wi-Fi, is chosen as the target standard. An interesting question that has practical implications is whether there is any benefit in using more than one wireless technology to perform the localization. In this paper we present a study on the advantages and challenges of using multiple wireless technologies to perform localization in indoor environments. We use real ZigBee, Wi-Fi and Bluetooth compliant devices. In our study we analyse results obtained using the fingerprint method. The performance of each technology alone and the performance of the technologies combined are also investigated. We also analyse how the number of wireless devices used affects the quality of localization and show that, for all technologies, more beacons lead to less error. Finally, we show how interference among technologies may lead to lower localization accuracy.

There are several applications for which it is important to both detect and communicate changes in data models. For instance, in some mobile robotics applications (e.g. surveillance) a robot needs to detect significant changes in the environment (e.g. a layout change) which it may achieve by comparing current data provided by its sensors with previously acquired data (e.g. map) of the environment. This often constitutes an extremely challenging task due to the large amounts of data that must be compared in real-time. This paper proposes a framework to detect, and represent changes through a compact model. The main steps of the procedure are: multi-scale sampling to reduce the computation burden; change detection based on Gaussian mixture models; fitting superquadrics to detected changes; and refinement and optimization using the split and merge paradigm. Experimental results in various real and simulated scenarios demonstrate the approach's feasibility and robustness with large datasets.

This paper presents a methodology based on a variation of the Rapidly-exploring Random Trees (RRTs) that generates feasible trajectories for autonomous vehicles with holonomic constraints in environments with obstacles. Our approach is based on seventh order Bézier curves to connect vertexes of the tree, generating paths that do not violate the main kinematic constraints of the vehicle. The methodology also does not require complex kinematic and dynamic models of the vehicle. The smoothness of the acceleration profile of the entire path is directly guaranteed by controlling the curvature values at the extreme points of each Bézier that composes the tree. The proposed algorithm provides fast convergence to the final result with several other advantages, such as the reduction in the number of vertexes of the tree because the method enable connections between vertexes of the tree with unlimited range. In an environment with few obstacles, a very small quantity of vertexes (sometimes only two) is sufficient to take the robot between two points. The properties of the seventh order Bézier formulation are also used to avoid collisions with static obstacles in the environment.

The ability to detect perceptions which were never experienced before, i.e. novelty detection, is an important component of autonomous robots working in real environments. It is achieved by comparing current data provided by its sensors with a previously known map of the environment. This often constitutes an extremely challenging task due to the large amounts of data that must be compared in real-time. With respect to previously proposed approaches, this paper detects changes in 3D environment based on probabilistic models, the Gaussian Mixture Model, and a fast and robust combined constraint matching algorithm. The matching allows to represent the scene view as a graph which emerges from the comparison between Mixtures of Gaussians. Finding the largest set of mutually consistent matches is equivalent to find the maximum clique on a graph. The proposed approach has been tested for mobile robotics purposes in real environments and compared to other matching algorithms. Experimental results demonstrate the performance of the proposal.

This paper presents a methodology based on a variation of the quintic Pythagorean hodographs curves for generating smooth feasible paths for autonomous vehicles in three-dimensional space under the restriction of limited climb angles. A given path is considered feasible if the main kinematic constraints of the vehicle are not violated. The generated paths satisfy three main angular constraints: (i) maximum curvature, (ii) maximum torsion and (iii) maximum climb (or dive). The smoothness the vehicle acceleration profile is indirectly guaranteed between two consecutive points of the profile. The proposed methodology is applicable to vehicles that move in three-dimensional environments, and that can be modelled under the constraints assumed. We apply our methodology and show the results for a small autonomous aerial vehicle.

This paper presents a methodology based on a variation of the Rapidly-exploring Random Trees (RRTs) that generates feasible trajectories for autonomous aerial vehicles with holonomic constraints in environments with obstacles. Our approach uses Pythagorean Hodograph (PH) curves to connect vertices of the tree, which makes it possible to generate paths for which the main kinematic constraints of the vehicle are not violated. The smoothness of the acceleration profile of the vehicle is indirectly guaranteed between two vertices of the RRT tree. The proposed algorithm provides fast convergence to the final trajectory. We still utilize the properties of the RRT to avoid collisions with static obstacles of a environment. We show results for a small unmanned aerial vehicles in environments with different configurations.

In this paper we present a methodology based on a variation of the spatial pythagorean hodograph curves to generate smooth feasible paths for autonomous vehicles in three-dimensional space under the restriction of limited climb angles. An fast iterative algorithm is used to calculate the curve. The generated path satisfy three main angular constraints given by the vehicle: (i) maximum curvature, (ii) maximum torsion and (iii) maximum climb (or dive). A path is considered feasible if these kinematic constraints are not violated. The smoothness vehicle's acceleration profile is indirectly guaranteed between two points. The proposed methodology is applicable to vehicles that move in three-dimensional environments, and that can be modeled by the constraints considered here. We show results for small aerial vehicle.

This paper describes an efficient method for retrieving the 3-dimensional shape associated to novelties in the environment of an autonomous robot, which is equipped with a laser range finder. First, changes are detected over the point clouds using a combination of theGaussian mixture model(GMM) and theearth mover's distance(EMD) algorithms. Next, the shape retrieval is achieved using two different algorithms. First, new samplings are generated from each Gaussian function, followed by arandom sampling consensus(RANSAC) algorithm to retrieve geometric primitives. Furthermore, a new algorithm is developed to directly retrieve the shape according to the mathematical space of Gaussian mixture. In this paper, the set of geometric primitives has been limited to the setC = {sphere, cylinder, plane}. The two shape retrieval methods are compared in terms of computational cost and accuracy. Experimental results in various real and simulated scenarios demonstrate the feasibility of the approach.

We propose and implement a hybrid navigational strategy for mobile robots for data collection tasks in sparsely deployed sensor networks. Our approach presents an alternative to the routing algorithms used in wireless sensor networks which focus on the minimization of energy consumption of each node which is mainly due to data transmission. The method has two layers: i) Reactive Layer in which the data collected from a sensor node is modelled as a potential function whose gradient attracts the robot and ii) Planning Layer, that dictates the action the robot must take to collect sensor data. Results from both simulation and real world experiments showing the performance of the proposed methodology are shown, compared and discussed.

This paper describes the derivation of dynamical models for mobile robots using system-identification methods and shows how such models can be used in the control of autonomous devices. Unlike traditional approaches that are based on the system physics alone, in this paper, an autoregressive model structure is derived from the physics of the process, and its parameters are estimated using real input and output data. From the resulting model, efficient control can be obtained for centralized and remotely controlled robots (typically externally observed) with little or no local "intelligence," thus overcoming many of the severe shortcomings presented by such systems. System dead time is one example of a serious problem that can be efficaciously minimized by using the model as an online predictor. Several experimental results using small differential-driven robots are included to demonstrate the applicability and robustness of the suggested procedure in actual real-world scenarios

This paper addresses the problem of efficiently computing robot navigation functions. Navigation functions are potential functions free of spurious local minima that present an exact solution to the robot motion planning and control problem. Although some methodologies were found in the literature, none of them are easy to implement and generalize for complex shaped workspaces and robots. We discuss some of the difficulties encountered in the current methodologies and propose a novel approach using a Finite Element method for potential field computation.

This paper addresses the problem of collecting sensoring information from wireless sensor networks using mobile robots. We consider that sensors may interact with each other only to maintain the network topology. Because transmitting sensoring data over the network to a remote base is a very power consuming operation, we deploy mobile robots through the sensor field in order to collect such information. We present a multi-target, dynamical, potential field strategy where a robot is attracted to the sensors that have less remaining data storage capacity. Moreover, the robot is also attracted to other sensors in its path without compromising its principal mission. Our controllers guarantee the robot will never get stuck in undesirable local minima. Simulated examples with a single robot collecting data in a sensor field illustrate the elegancy of the methodology.

This paper proposes a methodology that uses hybrid systems to model multiple robots in the execution of cooperative tasks. Basically, each robot is represented by a hybrid automation and the cooperative task execution is modeled by the composition of several automata. We describe in details our approach to perform the composition of these automata and demonstrate the effectiveness of the proposed methodology modeling a cooperative manipulation task.

We address the problem of controlling a team of robots subject to constraints on relative positions. We adopt the general framework of leader-follower control in which a network of controllers is used to control the position and orientation of the team and its shape. We propose two improvements to this scheme. First, we introduce cooperative leader-following where the motion of a robot is determined not only by its leader, but also by other robots including their followers. Second, we allow constraints that are induced by limitations on ranges of sensors and wireless network cards. Our approach is based on potential field controllers for each robot and the on-line modification of these controllers to accommodate motion constraints induced by other robots in the group. We present experimental results with a team of three car-like robots equipped with omnidirectional cameras and 802.11b network cards.

This work addresses the problem of multi-mobile robot cooperation with strict communication constraints which are considered indispensable for successful task execution. The problem, which is instantiated as a cooperative search and rescue task, is modeled as a minimization of an energy functional which accounts for network connectivity and other relevant robot and task requirements in order to select locally optimal actions for each robot. The model is tested against several scenarios and the results produce elegant and rather efficient solutions to the proposed problem, albeit not globally optimal.

A semi-autonomous robot for autonomous installation and removal of aircraft warning spheres on overhead wires of electric power transmission lines is described. Hybrid systems methodology was used to model the basic modes of operation of the robot, which includes both autonomous and tele-operation possibilities. Several sensors and actuators including a custom made gripper are used to accomplish the demanded tasks. The robot was built and tested in real operational situations and proved to be both efficient and robust. It will enable power distribution companies to replace current highly dangerous operations which are normally performed by a human operator placed in a helicopter hovering a few meters away of hotlines. Broader use of this robot in several other common tasks such as line inspection and cleaning, cable changing and object transportation are discussed.

Proposes a methodology for coordinating multi-robot teams in the execution of cooperative tasks. It is based on a dynamic role-assignment mechanism in which the robots assume and exchange roles during cooperation. We model the role assignment under a hybrid systems framework, using a hybrid automaton to represent roles, transitions and controllers. Using a multi-robot simulator, the methodology is demonstrated in a cooperative transportation task, in which a group of robots must find and cooperatively transport several objects scattered in the environment.

Addresses the problem of coordinating multiple mobile robots in a tightly coupled task by means of implicit communication. This approach allows the development of controllers that do not depend on any explicit data flow between the robots, thus relying only on local sensor information. A box-carrying task is used to validate the proposed methodology both in simulation and in real-world experiments. Results show that implicit communication can be used together or replacing explicit communication for the cooperative box carrying task, making the system more robust to faulty communication environments.

Proposes an architecture for tightly coupled multi-robot coordination that is well suited to cooperative manipulation tasks. At all times, a robot is identified as a leader, while the others are designated as followers. The assignment of roles and the coordination between the robots is guaranteed by communication protocols and control algorithms. The key feature is the flexibility that allows changes in leadership and assignment of roles during the execution of a task. We describe the experimental implementation and demonstration in a cooperative transportation task, in which two and three heterogeneous robots cooperate to carry a large object in an environment containing obstacles.

Recent investigation on the cooperation of multiple robotic agents with centralized control, has ushered in small and less sophisticated robots that are observed by a global vision system. This paper presents a modeling methodology that enables accurate prediction of such small robot's position and orientation from data measured using the global vision system. Results of several experiments show that the predictions of the model can be used to compensate time delays and improve the transient response of the robots.

Until recently the control and navigation of mobile robots was mainly performed based on exteroceptive sensoring. Recent investigation on the cooperation of multiple robotic agents with centralized control, has ushered in small and less sophisticated robots that are observed by a global vision system. This paper presents a modeling methodology that enables accurate prediction of such small robot's position and orientation from data measured using the global vision system. Results of several experiments show that the model is both accurate and robust. Another relevant aspect of the work is the ability to determine the limitations likely to be found in centralized control of agents with little or no local "intelligence".

Describes a methodology for visual navigation of dirigibles based on tracking and pose estimation of visual beacons-artificial landmarks with known geometrical and visual properties. Fast pose estimation for the dirigible is necessary since the 3D position and orientation data may be used for attitude control. Robust tracking of the beacons is also an important feature. The methodology is validated through different experiments. Results on precision of position and velocity parameters, robustness in tracking the beacon in unstructured environments and time performance of the system are discussed. Finally, future improvements and applications to external airships are discussed.