Search

The robot behavior strategy is considered as a crucial part in the human-following task to help the robot maintain an appropriate distance and orientation to the selected target person (STP) with a smooth and safe manner. As usual, the robot is uniquely considered to follow the STP in a specific class of environments, such as unknown environments (non-mapped environments) or known environments (mapped environments). However, in real-life applications, the robot is sometimes requested to follow the STP in various types of environments, both in known and unknown ones. This observation raises the need to propose an alternative method to challenge the mentioned issue, as well as to break the current limit of the human-following function. In this paper, a new approach for the human-following strategy is proposed in which the mobile robot is enabled to follow the STP in mixed environments (non-mapped and mapped). In non-mapped environments, only the STP and the obstacle information with respect to the robot local coordinates are considered, whose purpose is to make the robot work without any prior understandings about its working environment. However, after the robot entered mapped environments, its prior knowledge of the working environment is leveraged to fulfill some additional requirements during the cooperation, such as the mobile robot in factories is not allowed to enter some specific areas even when the STP is executing technical tasks inside. Additionally, in this paper, a human-like inference mechanism is also introduced for the human-following strategy by using an extended hedge algebras. The proposed method is experimentally verified both in factories and laboratories. Demo Video Link : https://www.youtube.com/watch?v=YGrWU6ldKuw Since real videos in the factory are not allowed to publish, only visualization (in Rviz) is presented for demos in such kinds of environments. The visualization is synchronous with the real executions of the human–robot interactions. The robot used in the factory is an autonomous mobile robot (dimension 0.5 (m) ×1.0 (m), weight 120 (kg), carrying a tool cabinet around 300(kg))). The mobile robot is following the worker to support them during the technical processes in the car production line. In the video, the robot is represented by a green rectangular, and the STP is represented by a cylinder (with a sphere on its head) The events in the demo video are described more clearly in Appendix A. • First contribution: a new concept for human-following robots in mixed environments (mapped and non-mapped). • Second contribution: a hedge-algebras-based approach for the robot behavior strategy to work both in indoor and outdoor, mapped and non-mapped environments, and in different robot platforms. • Third contribution: an end-to-end human-following system for mobile robots by integrating all required modules, from trigger methods, the human detection and tracking to robot behavior policies and the inference mechanism. • Experiments are conducted both in factories and laboratories. And obtained results have shown that the proposed method works well in various environments, both indoor and outdoor and ondifferent robot platforms. [ABSTRACT FROM AUTHOR]

Confined and subterranean areas are common in many civilian and industrial sites, although they are hazardous for humans given the presence of noxious gases, extreme temperatures, narrow spaces, unhealthy oxygen levels, flooding, and collapsing structures. Therefore, exploration, routine inspections, and surveillance tasks can benefit from using autonomous mobile robots to improve safety by reducing the presence of humans in those scenarios. However, despite advances in the field, there are still challenges to overcome for confined and subterranean robot operation. Real word robotic exploration requires robust and reliable map generation, precise localization, safe navigation, and efficient path planning. These requirements make exploration in complex 3D environments with rugged terrain difficult. The challenge is increased when considering multi-robot teams, as there is no guarantee of a functional network infrastructure. Despite consistent increasing interest in the area, there is a lack of research summarizing the results and best practices for exploring such environments. Therefore, in this paper, we provide a review and discuss state-of-the-art robotic exploration techniques, including single and cooperative approaches with homogeneous and heterogeneous teams, with a focus on complex subterranean and confined 3D scenarios. We also present a comprehensive list of insights on open challenges and possible directions for future investigation in the topic. • A survey in autonomous exploration for subterranean and confined spaces using mobile robots. • Description of current state of the art, including results from the DARPA subt Challenge. • Sensor classification for confined spaces inspection. • Comparison of recent works in autonomous exploration using multi-robot systems. [ABSTRACT FROM AUTHOR]

As for the uncertainty problem of detection and estimation of robot localization and mapping using ultrasonic sensor, this paper proposes the improvement of Strong Tracking UKF-SLAM approach using three-position ultrasonic detection. A three-position ultrasonic detection model is first of all built for reducing these uncertainties through topological relationship screening and environmental contour estimation. Then Strong Tracking UKF-SLAM approach is improved by using multiple fade factors to fuse the ultrasonic measurement data and motion model information of robot for obtaining more accurate localization and mapping. Finally, we construct simulation and indoor experimental environments and design the mobile robot system with ultrasonic sensor for verification. The simulation represents that the improved algorithm has less error and more accurate effect than original algorithms in localization and mapping of mobile robot. The indoor environmental experiment is performed for illustrating the feasibility and effectiveness of the proposed method. The proposed method has certain reference value for research of Simultaneous Localization and Mapping. • Focus on simultaneous localization and mapping of mobile robot in complex indoor environments using ultrasonic sensor. • Establishing arc based measurement model of ultrasonic sensors for filtering noise data. • Mapping environment by using three adjacent position ultrasonic detection model. • Improvement of strong tracking UKF-SLAM algorithm with more accuracy. [ABSTRACT FROM AUTHOR]

To navigate and control a single mobile robot or a robotic swarm with higher efficiency, a novel non-potential orthogonal vector field method is proposed in this paper, which is modified from the traditional artificial potential field method. The improvement strategy aims at making the overall repulsive vector field orthogonal to the attractive vector field in some conditions. And the same potential field function is still applied to the Lyapunov-based stability analysis. In short, such an improvement strategy combines the advantages of the artificial potential field method with the non-potential vector field method, namely a combination of theoretical completeness and control efficiency. Finally, the effectiveness of the proposed method is validated both by numerical simulations with statistical significance and real experiments. The comparisons between our method and other methods are also presented. [ABSTRACT FROM AUTHOR]

Autonomous mobile robots are a special class of robotic systems that can move a payload from one location to the other or perform a specific task. They allow efficient, precise, and streamlined workflow that makes human work less arduous. The market and research work related to these robots is increasing in anticipation of industry 5.0, where humans and machines are expected to co-exist and co-work. The future mobile robots are desired to have clean and cost-effective energy sources to have longer operation times and compliance with environmental requirements to allow application in diverse fields. The research on mechanical design, perception, navigation and control has carved out many commercially viable solutions for mobile robots. However, their widespread application is still limited due to the lack of efficient power systems for use in diverse and largely unknown/uncontrolled environments. The current power solutions incur high initial costs and require recharging or refuelling, which makes them unsuitable for unattended long-haul worktimes and cost-effective applications. These drawbacks are major hurdles in the wider applicability of terrain-based mobile robots to new domains and daily life scenarios, which are possible with the existing mechanical, perception, and control technologies. Keeping in view the need for advancement in this field and to gain a better understanding of the current state of the art and future directions, this work summarizes and reviews the energy solutions presented in the literature and used in notable commercially available terrain-based mobile robots. The provided solutions are categorized and discussed while the prospects and research gaps are also highlighted. A comparison of discussed power techniques is also provided, which can serve as a guideline for selecting a robot's energy source according to the desired requirements. [ABSTRACT FROM AUTHOR]

The availability of maps of indoor environments is often fundamental for autonomous mobile robots to efficiently operate in industrial, office, and domestic applications. When robots build such maps, some areas of interest could be inaccessible, for instance, due to closed doors. As a consequence, these areas are not represented in the maps, possibly causing limitations in robot localization and navigation. In this paper, we provide a method that completes 2D grid maps by adding the predicted layout of the rooms behind closed doors. The main idea of our approach is to exploit the underlying geometrical structure of indoor environments to estimate the shape of unobserved rooms. Results show that our method is accurate in completing maps also when large portions of environments cannot be accessed by the robot during map building. We experimentally validate the quality of the completed maps by using them to perform path planning tasks. • Doors that are closed at mapping time could make 2D grid maps incomplete. • We predict the layout of closed rooms using geometrical structural features. • Experiments show accurate predictions and an application to path planning. [ABSTRACT FROM AUTHOR]

This paper presents a robust intelligent controller to be applied to a class of nonholonomic electrically driven mobile robots. This class of robotic systems has an inherent sensitivity to high degree time-varying parametric uncertainties, unmodeled dynamics, and external disturbances. Furthermore, the effects of coupling terms between the mechanical subsystem and the electrical subsystem may cause severe degradations due to the time-varying variations of DC motors and mechanical structure components around their nominal values. To overcome the effects of all these quantities, the robust adaptive neural network tracking controller developed here introduces adaptive laws to estimate a local upper bound of each subsystem of the nonholonomic mobile robot, then, these laws are used on-line as controller gain parameters in order to robustly improve the transient response of the closed-loop system and reduce conservative, in the sense that the local upper bounds to characterize the corresponding uncertainties dynamics for each subsystem, initially computed based on the worse-case scenario, are not updated during the effective control of the mobile robot. In fact, even if more data become available, then they are avoided when estimating local upper bounds, and hence, the level of uncertainty is considerably decreased. According to the universal approximation theorem and the Lyapunov stability theory, the proposed intelligent controller guarantees global stability in the sense that all the states and signals of the closed-loop system, and the trajectory tracking errors are all bounded. Simulation results on two typical examples of nonholonomic electrically driven mobile robots show the effectiveness and robustness of the proposed controller. [ABSTRACT FROM AUTHOR]

Recognizing semantic category of objects and scenes captured using vision-based sensors is a challenging yet essential capability for mobile robots and UAVs to perform high-level tasks such as long-term autonomous navigation. However, extracting discriminative features from multi-modal inputs, such as RGB-D images, in a unified manner is non-trivial given the heterogeneous nature of the modalities. We propose a deep network which seeks to construct a joint and shared multi-modal representation through bilinearly combining the convolutional neural network (CNN) streams of the RGB and depth channels. This technique motivates bilateral transfer learning between the modalities by taking the outer product of each feature extractor output. Furthermore, we devise a technique for multi-scale feature abstraction using deeply supervised branches which are connected to all convolutional layers of the multi-stream CNN. We show that end-to-end learning of the network is feasible even with a limited amount of training data and the trained network generalizes across different datasets and applications. Experimental evaluations on benchmark RGB-D object and scene categorization datasets show that the proposed technique consistently outperforms state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]

Crossing surface discontinuities, such as ditches or curbs in urban environments, is a relevant navigation issue for ground mobile robots. This paper proposes a maneuver for negotiating surface discontinuities for a hybrid locomotion system consisting on a skid-steer wheeled vehicle with a single caster-leg mechanism. This maneuver is appropriate for gaps, steps, and ramps, both upwards and downwards. Furthermore, an analysis of tipover stability and steerability through the different states of the maneuver is performed, resulting in sensor-based stability margins that can be computed online. Experimental results have been obtained by means of realistic Simulink/ADAMS simulations and actual tests with the Lazaro mobile robot. The results confirm the effectiveness of the proposed maneuver as well as the applicability of stability margins. [ABSTRACT FROM AUTHOR]

This paper provides a novel distributed predictive controller with guaranteed stability to maintain formation between mobile robots during their motion along a desired path and assure no collisions with obstacles or other adjacent robots, while data is exchanged among them via a packet-delaying communication network. First, the closed-loop system dynamics are described as a delayed differential equation with tunable parameters. Then, these adjustable gains are determined synchronously in each agent by the proposed predictive strategy such that a desirable formation is achieved. The efficiency and applicability of the suggested scheme are demonstrated by simulation results. [ABSTRACT FROM AUTHOR]

This paper presents an approach to developing an assistive interface for human–robot interaction that provides users with customized haptic and visual feedback. The developed interface targets to improve users’ task performance by customizing assistance policy and level based on users’ performance and control strategy. To achieve this, the users’ control strategy was modeled based on inverse optimal control technique. Then, features describing the geometric and behavioral characteristics of user control are derived. Finally, an expert whose features most closely matched each user was identified. The identified expert was assigned to the user to define and provide customized assistance via a virtual fixturing. In human subject experiments, control strategies of twenty-three users were identified and featured; their performance was measured with four assistance types (no-assist, haptic assistance, visual zooming assistance, haptic assistance + visual zooming assistance) × two parameterization types (customized, non-customized). By analyzing the experimental data, we found an optimal combination of assistance type × parameterization type that results in the most improvement of performance. The results showed that the users’ task completion time and mean required effort yielded the best improvements when haptic assistance (with no visual zooming assistance) × customized parameterization were provided for mobile robot driving tasks. [ABSTRACT FROM AUTHOR]

Multi-robot conflict resolution is a challenging problem, especially in dense environments where many robots must operate safely in a confined space. Centralized solutions do not scale well with the number of robots in dynamic scenarios: a centralized communication can cause bottlenecks and may not be robust enough when channels are unreliable; the complexity of algorithms grows with the number of robots, making online re-computation too expensive in many situations. In this work, we propose a decentralized approach for conflict resolution where robots show reactive and safe behaviors, avoiding collisions with both static and dynamic objects, even under unreliable communication conditions and with low resources. They detect conflicts with neighboring obstacles locally and then apply rules to surround them in a roundabout fashion, assuming that others will follow the same policy. The method is designed for unicycle robots with range-finder sensors, and it is able to cope with noisy sensors and second-order dynamic constraints, ensuring always collision-free navigation. Besides, a set of metrics and scenarios for benchmarking in multi-robot collision avoidance are proposed. We also compare our method with others from the state of the art through extensive simulations. Experiments with real robots are also presented in order to show the feasibility of the system. [ABSTRACT FROM AUTHOR]

In this paper, we propose a nonparametric motion controller using Gaussian process regression for autonomous navigation in a dynamic environment. Particularly, we focus on its applicability to low-cost mobile robot platforms with low-performance processors. The proposed motion controller predicts future trajectories of pedestrians using the partially-observable egocentric view of a robot and controls a robot using both observed and predicted trajectories. Furthermore, a hierarchical motion controller is proposed by dividing the controller into multiple sub-controllers using a mixture-of-experts framework to further alleviate the computational cost. We also derive an efficient method to approximate the upper bound of the learning curve of Gaussian process regression, which can be used to determine the required number of training samples for the desired performance. The performance of the proposed method is extensively evaluated in simulations and validated experimentally using a Pioneer 3DX mobile robot with two Microsoft Kinect sensors. In particular, the proposed baseline and hierarchical motion controllers show over 65 % and 51 % improvements over a reactive planner and predictive vector field histogram, respectively, in terms of the collision rate. [ABSTRACT FROM AUTHOR]

This paper presents preliminary results of the application of two-Kinect cameras system on a two-wheeled indoor mobile robot for off-line optimal path planning and execution. In our approach, the robot makes use of depth information delivered by the vision system to accurately model its surrounding environment through image processing techniques. In addition, a Genetic Algorithm is implemented to generate a collision-free optimal path linking an initial configuration of the mobile robot ( Source ) to a final configuration ( Target ). After that, Piecewise Cubic Hermite Interpolating Polynomial is used to smooth the generated optimal path. Finally, an Adaptive Fuzzy-Logic controller is designed to keep track of a mobile robot on the desired smoothed path (by transmitting the appropriate right and left velocities using wireless communication). In parallel, sensor fusion (odometry sensors and Kinect sensors) is used to estimate the current position and orientation of the robot using Kalman filter . The validation of the proposed solution is carried out using the differentially-driven mobile robot, RobuTER , to successfully achieve safe motion (without colliding with obstacles) in an indoor environment. [ABSTRACT FROM AUTHOR]

Multisensor perception has an important role in robotics and autonomous systems, providing inputs for critical functions including obstacle detection and localization. It is starting to appear in critical applications such as drones and ADASs (Advanced Driver Assistance Systems). However, this kind of complex system is difficult to validate comprehensively. In this paper we look at multisensor perception systems in relation to an alternative dependability method, namely fault tolerance. We propose an approach for tolerating faults in multisensor data fusion that is based on the more traditional method of duplication–comparison, and that offers detection and recovery services. We detail an example implementation using Kalman filter data fusion for mobile robot localization. We demonstrate its effectiveness in this case study using real data and fault injection. [ABSTRACT FROM AUTHOR]

We propose a hybrid approach specifically adapted to deal with the autonomous-navigation problem of a mobile robot which is subjected to perform an emergency task in a partially-known environment. Such a navigation problem requires a method that is able to yield a fast execution time, under constraints on the capacity of the robot and on known/unknown obstacles, and that is sufficiently flexible to deal with errors in the known parts of the environment (unexpected obstacles). Our proposal includes an off-line task-independent preprocessing phase, which is applied just once for a given robot in a given environment. Its purpose is to build, within the known zones, a roadmap of near-time-optimal reference trajectories. The actual execution of the task is an online process that combines reactive navigation with trajectory tracking and that includes smooth transitions between these two modes of navigation. Controllers used are fuzzy-inference systems. Both simulation and experimental results are presented to test the performance of the proposed hybrid approach. Obtained results demonstrate the ability of our proposal to handle unexpected obstacles and to accomplish navigation tasks in relatively complex environments. The results also show that, thanks to its time-optimal-trajectory planning, our proposal is well adapted to emergency tasks as it is able to achieve shorter execution times, compared to other waypoint-navigation methods that rely on optimal-path planning. [ABSTRACT FROM AUTHOR]

Many robots are limited in their operating capabilities, both computational and energy-wise. A strong desire exists to keep computation cost and energy consumption to a minimum when executing tasks like object recognition with a mobile robot. Adaptive action selection is a paradigm, offering great flexibility in trading off the cost of acquiring information against making robust and reliable inference under uncertainty. In this paper, we study active multi-view object recognition and describe an information-theoretic framework that combines and unifies two common techniques: online feature selection for reducing computational costs and view planning for resolving ambiguities and occlusions. Our algorithm adaptively chooses between the two strategies of either selecting only the features that are most informative to the recognition, or moving to a new viewpoint that optimally reduces the expected uncertainty on the identity of the object. This two step process allows us to keep overall computation cost minimal but simultaneously increase recognition accuracy. Extensive empirical studies on a large RGB-D dataset, and with two different feature sets, have validated the effectiveness of the proposed framework. Our experiments show that dynamic feature selection alone reduces the computation time at runtime 2.5–6 times and, when combining it with viewpoint selection, we significantly increase the recognition accuracy on average by 8%–18% absolute compared to systems that do not use these two strategies. By establishing a link between active object recognition and change detection, we were further able to use our framework for the follow-up task of actively detecting object change. Furthermore, we have successfully demonstrated the framework’s applicability to a low-powered quadcopter platform with limited operating time. [ABSTRACT FROM AUTHOR]

Autonomous vehicles have a large number of on-board sensors, not only for providing coverage all around the vehicle, but also to ensure multi-modality in the observation of the scene. Because of this, it is not trivial to come up with a single, unique representation that feeds from the data given by all these sensors. We propose an algorithm which is capable of mapping texture collected from vision based sensors onto a geometric description of the scenario constructed from data provided by 3D sensors. The algorithm uses a constrained Delaunay triangulation to produce a mesh which is updated using a specially devised sequence of operations. These enforce a partial configuration of the mesh that avoids bad quality textures and ensures that there are no gaps in the texture. Results show that this algorithm is capable of producing fine quality textures. [ABSTRACT FROM AUTHOR]

We introduce a novel measure of terrain unevenness, which is computed in terms of ranges of neighbouring laser beams of a 3D laser scanner mounted on a mobile robot. The unevenness so computed over all sampled points forms an unevenness field around the robot. We explore the nature of this measure of unevenness through analysis, and arrive at a reasonable policy for setting thresholds on unevenness in order to detect obstacles, and in the process mark out the traversable region around the mobile robot. The traversable region is obtained as a connectivity graph over 180 × 32 cells in about 30 ms time on an Intel i3 processor. This connectivity graph can be potentially used for path planning of the mobile robot. Conceptually and computationally the measure is simple and efficient. It has an added advantage of being robust against small tilts of the sensor during locomotion. It also works well on slopes. Although there are numerous ways of detecting traversable regions in the literature, it is the novelty of the measure and its analysis, which we believe is our contribution through this paper. We demonstrate its usefulness through experimental results and compare its performance with a standard method that uses both height and slope between neighbouring range points to detect obstacles. [ABSTRACT FROM AUTHOR]

Designing intelligent and robust autonomous navigation systems remains a great challenge in mobile robotics. Inverse reinforcement learning (IRL) offers an efficient learning technique from expert demonstrations to teach robots how to perform specific tasks without manually specifying the reward function. Most of existing IRL algorithms assume the expert policy to be optimal and deterministic, and are applied to experiments with relatively small-size state spaces. However, in autonomous navigation tasks, the state spaces are frequently large and demonstrations can hardly visit all the states. Meanwhile the expert policy may be non-optimal and stochastic. In this paper, we focus on IRL with large-scale and high-dimensional state spaces by introducing the neural network to generalize the expert’s behaviors to unvisited regions of the state space and an explicit policy representation is easily expressed by neural network, even for the stochastic expert policy. An efficient and convenient algorithm, Neural Inverse Reinforcement Learning (NIRL), is proposed. Experimental results on simulated autonomous navigation tasks show that a mobile robot using our approach can successfully navigate to the target position without colliding with unpredicted obstacles, largely reduce the learning time, and has a good generalization performance on undemonstrated states. Hence prove the robot intelligence of autonomous navigation transplanted from limited demonstrations to completely unknown tasks. [ABSTRACT FROM AUTHOR]

This paper presents a novel reactive collision avoidance method for mobile robots moving in dense and cluttered environments. The proposed method, entitled Tangential Gap flow (TGF), simplifies the navigation problem using a divide and conquer strategy inspired by the well-known Nearness-Diagram Navigation (ND) techniques. At each control cycle, the TGF extracts free openings surrounding the robot and identifies the suitable heading which makes the best progress towards the goal. This heading is then adjusted to avoid the risk of collision with nearby obstacles based on two concepts namely, tangential and gap flow navigation. The tangential navigation steers the robot parallel to the boundary of the closest obstacle while still emphasizing the progress towards the goal. The gap flow navigation safely and smoothly drives the robot towards the free area in between obstacles that lead to the target. The resultant trajectory is faster, shorter and less-oscillatory when compared to the ND methods. Furthermore, identifying the avoidance maneuver is extended to consider all nearby obstacle points and generate an avoidance rule applicable for all obstacle configurations. Consequently, a smoother yet much more stable behavior is achieved. The stability of the motion controller, that guides the robot towards the desired goal, is proved in the Lyapunov sense. Experimental results including a performance evaluation in very dense and complex environments demonstrate the power of the proposed approach. Additionally, a discussion and comparison with existing Nearness-Diagram Navigation variants is presented. [ABSTRACT FROM AUTHOR]

This paper describes a person identification method for mobile service robots using image and range data. Person identification is a necessary function in order for mobile service robots to locate the target person for those services. Among various sensory features, image-based appearance features have often been used for person identification. They are, however, not effective in severe illumination environments such as a strong backlight. Therefore, we use two illumination-independent features, height and gait, in addition to appearance features for a more robust identification. To this end, we have developed a new method of extracting the gait feature (step length and speed), based on a maximum likelihood estimation of supporting leg positions in accumulated range data. We combine these features and use an online boosting approach to create the specific person classifier. It allows the robot to identify the specific person robustly even in a severe illumination environment. We tested our multi-feature person identification method, combined with a range data-based person tracker, in a specific person following scenario to demonstrate the effectiveness of this method. [ABSTRACT FROM AUTHOR]

Localization of mobile robots is still an important topic, especially in case of dynamically changing, complex environments such as in Urban Search & Rescue (USAR). In this paper we aim for improving the reliability and precision of localization of our multimodal data fusion algorithm. Multimodal data fusion requires resolving several issues such as significantly different sampling frequencies of the individual modalities. We compare our proposed solution with the well-proven and popular Rauch–Tung–Striebel smoother for the Extended Kalman filter. Furthermore, we improve the precision of our data fusion by incorporating scale estimation for the visual modality. [ABSTRACT FROM AUTHOR]

We address the problem of controlling a mobile robot to explore a partially known environment. The robot’s objective is the maximization of the amount of information collected about the environment. We formulate the problem as a partially observable Markov decision process (POMDP) with an information-theoretic objective function, and solve it applying forward simulation algorithms with an open-loop approximation. We present a new sample-based approximation for mutual information useful in mobile robotics. The approximation can be seamlessly integrated with forward simulation planning algorithms. We investigate the usefulness of POMDP based planning for exploration, and to alleviate some of its weaknesses propose a combination with frontier based exploration. Experimental results in simulated and real environments show that, depending on the environment, applying POMDP based planning for exploration can improve performance over frontier exploration. [ABSTRACT FROM AUTHOR]

This paper presents a concept for path planning of mobile robots in household environments. The proposed algorithm takes communicative means of locomotion into account in order to facilitate human–robot interaction. The well-known traditional artificial potential field method (APF) was extended by motion characteristics of household animals. The proposed algorithm includes velocity and orientation information and can be used in unknown, dynamic environments. The main contribution of this paper is the definition of an online, local path planning method by adapting animal motion attributes in order to assist human–robot interaction. The algorithm was implemented in an embedded system and evaluated on MOGI-ETHON, a holonomic drive mobile robot. [ABSTRACT FROM AUTHOR]

This paper presents a recurrent neural network based novelty filter where a Scitos G5 mobile robot explored the environment and built dynamic models of observed sensory–motor values, then the acquired models of normality are used to predict the expected future values of sensory–motor inputs during patrol. Novelties could be detected whenever the prediction error between models-predicted values and actual observed values exceeded a local novelty threshold. The network is trained on-line; it grows by inserting new nodes when abnormal observation is perceived from the environment; and also shrinks when the learned information is not necessary anymore. In addition, the network is also capable of learning region-specific novelty thresholds on-line continuously. To evaluate the proposed algorithm, real-world robotic experiments were conducted by fusing sensory perceptions (vision and laser sensors) and the robot motor control outputs (translational and rotational velocities). Experimental results showed that all of the novelty cases were highlighted by the proposed algorithms and it produced reliable local novelty thresholds while the robot patrols in the noisy environment. The statistical analysis showed that there was a strong correlation between the novelty filter responses and the actual novelty status. Furthermore, the filter was also compared with another novelty filter and the results showed that the proposed system performed better novelty detection. [ABSTRACT FROM AUTHOR]

Robust environmental perception is a crucial parameter for the development of autonomous ground vehicle applications, especially in the field of agricultural robotics which is one of the priorities for the Horizon 2020 robotics funding (EU funding program for research and innovation). Because of uncontrolled and changing environmental conditions in outdoor and natural environments, data from optical sensors classically used in mobile robotics can be compromised and unusable. In such situations, millimeter-wave radar can provide an alternative and complementary solution for perception tasks. The aim of this paper is to present the PELICAN radar, a millimeter-wave radar specifically designed for mobile robotics applications, including obstacle detection, mapping and situational awareness in general. In this first of a two-part paper, the choice of a frequency-modulated continuous-wave radar is explained and the theoretical elements of this solution are detailed. PELICAN radar is using a rotating fan-beam antenna, and the construction of 2D representations of the surrounding environments with radar data is described through simulation results. The second part of the paper will be devoted for a detailed description of PELICAN radar, as well as experimental results . [ABSTRACT FROM AUTHOR]

Robotic calibration allows for the fusion of data from multiple sensors such as odometers, cameras etc., by providing appropriate transformational relationships between the corresponding reference frames. For wheeled robots equipped with exteroceptive sensors, calibration entails learning the motion model of the sensor or the robot in terms of the odometric data, and must generally be performed prior to performing tasks such as simultaneous localization and mapping (SLAM). Within this context, the current trend is to carry out simultaneous calibration of odometry and exteroceptive sensors without using additional hardware. Building upon the existing simultaneous calibration algorithms, we put forth a generalized calibration framework that can not only handle robots operating in 2D with arbitrary or unknown motion models but also handle outliers in an automated manner. We first propose an algorithm based on the alternating minimization framework applicable to two-wheel differential drive. Subsequently, for arbitrary but known drive configurations we put forth an iteratively re-weighted least squares methodology leveraging an intelligent weighing scheme. Different from the existing works, these proposed algorithms require no manual intervention and seamlessly handle outliers that arise due to both systematic and non-systematic errors. Finally, we put forward a novel Gaussian Process-based non-parametric approach for calibrating wheeled robots with unknown or un-modeled drive configurations. Detailed experiments are performed to demonstrate the accuracy, usefulness, and flexibility of the proposed algorithms. 1 1 This paper has supplementary downloadable material (available at http://tinyurl.com/simultaneous-calibration) that includes raw-data of all the experiments and implementation codes for the proposed methodologies. • A generalized framework for simultaneous calibration of wheeled robots. • Automatic outlier rejection mechanism. • Applicability to robots with arbitrary and/or unknown drive configurations. • Low-complexity approach for calibrating robots with unknown kinematics. • Extensive experiments in real settings. [ABSTRACT FROM AUTHOR]

The haptic terrain classification is an essential component of a mobile walking robot control system, ensuring proper gait adaptation to the changing environmental conditions. In practice, such components are a part of an autonomous system and thus have to be lightweight, provide fast inference time, and guarantee robustness to minor changes in recorded sensory data. We propose transformer-based HAPTR and HAPTR2 terrain classification methods that use force and torque measurements from feet to meet these requirements. For reliable comparison of the proposed solutions, we adapt two classical machine learning algorithms (DTW-KNN and ROCKET), one temporal convolution network (TCN), and use the state-of-the-art CNN-RNN. The experiments are performed on publicly available PUTany and QCAT datasets. We show that the proposed HAPTR and HAPTR2 methods achieve accuracy on par or better than state-of-the-art approaches with a lower number of parameters, faster inference time, and improved robustness to input signal distortions. These features make HAPTR and HAPTR2 excel in terrain recognition tasks when considering real-world requirements. • Fast and lightweight transformer-based approach for the terrain classification. • Modality Attention Layer improves the system's robustness in real-world scenarios. • Autonomy requirements inspired our evaluation of terrain classification algorithms. • The benchmark on two open source datasets creates a baseline for further research. [ABSTRACT FROM AUTHOR]

This study focuses on the problem of determining the posture of a mobile manipulator aiming to reach a target end-effector pose. Any posture of a mobile manipulator can be described using two parameters: the pose of the mobile platform and the joint angles of the manipulator. This study proposes a posture evaluation method for determining the pose of the mobile platform while focusing on two types of uncertainty: (i) the pose error associated with the end-effector caused by the accumulated positioning error of the mobile platform and (ii) the pose estimation error of objects to be grasped. In addition to using pose errors, the proposed method considers the manipulation ability of the end-effector and the tolerance region for grasping an object by the end-effector. Furthermore, this study proposes a method based on Bayes optimization for finding acceptable moving paths for the mobile platform in environments with sparse obstacles. The effectiveness of the proposed methods was demonstrated through simulations using a mobile manipulator. According to simulation results, the proposed methods can find both moving trajectories for the mobile platform and postures for the mobile manipulator within several seconds. • A method of approaching motion planning for mobile manipulators is proposed. • The method considers the uncertainty of both the robot's self-positioning and the pose estimation of a target object. • The proposed method also makes it possible to select a moving path considering obstacle avoidance. • The effectiveness of the proposed method is verified on the problem of grasping a parallelepiped box. [ABSTRACT FROM AUTHOR]

Most of the world's most advanced defense technologies are robots, and the defence industry is slowly moving toward including AI in the military robots they build. For these smart robots to make their own decisions about where to go and what to do, they need to be limited by several algorithms that run continuously and at the same time. Autonomy is the range of automated systems that can be adapted to a specific mission, residual risk, and level of team cohesion between humans and robots. Self-driving robotic systems should be collaborative, which means they should be able to interact actively with humans in a shared space or in proximity to humans and robots. Human–Robot Collaboration (HRC) works better when these COBOTs are aware of their surroundings. Mobile Robot (MR) teams whose perceptual and cognitive abilities are very well developed can help a lot with context awareness. To work well with humans, these robots should know what is going on with their human and other robot teammates so they can make decisions on their own. Also, robots should be able to share information about their surroundings so that humans can benefit from a better understanding of the situation. At the same time, humans should be able to see what the robots are doing. In this paper, we propose a knowledge-based framework for humans and robots to work together to understand the context of Defense missions. An ontological model of contexts for missions, agents, and situations; a knowledge base comprising all the tools necessary for a sort of situation; and an efficient and reliable method of collaborative learning are some of its main contributions. The framework works well in terms of how long it takes for people to talk to each other. As the team continues to expand, it can also easily manage communication challenges and a widely differing event frequency range. • We propose a knowledge-based framework for humans and robots to work together to understand the context of Defense missions. • By delivering this article, we intend to advance the concept of integrating CI into military agents. • The ontology-based Context-Aware Collaborative Agent distributes context patterns & improves the interaction and robot soldiers. • POMDP encompasses a conventional MDP with a HMM that deterministically links unobserved heterogeneous system states to findings. [ABSTRACT FROM AUTHOR]

Task and motion planning in robotics are typically addressed by separated intertwined methods. Task planners generate abstract high-level actions to be executed, while motion planners provide the associated discrete movements in the configuration space satisfying kinodynamic constraints. However, these two planning processes are strictly dependent, therefore the problem of combining task and motion planning with a uniform approach is very relevant. In this work, we tackle this issue by proposing a RRT-based method that addresses combined task and motion planning. Our approach relies on a combined metric space where both symbolic (task) and sub-symbolic (motion) spaces are represented. The associated notion of distance is then exploited by a RRT-based planner to generate a plan that includes both symbolic actions and feasible movements in the configuration space. The proposed method is assessed in several case studies provided by a real-world hospital logistic scenario, where an omni-directional mobile robot is involved in navigation and transportation tasks. • We propose an approach to combined task and motion planning based on RRTs. • The approach relies on a combined symbolic and sub-symbolic metric space. • A distance in the combined space is used by the RRT to plan actions and motions. • The method is assessed in case studies provided by a hospital logistic scenario. [ABSTRACT FROM AUTHOR]

Recently, vision-based methods have exhibited promising prospects in assigning autonomous robots more capabilities for better environment perception. Since visual sensors are easily affected under extreme conditions, current image inpainting methods based on CNNs jointly with generative adversarial networks (GANs) usually generate patches quite different from the ground truth (GT), which is harmful to autonomous robots. In this paper, we propose a novel multiscale feature alignment module with an early fusion strategy to align the left and right feature maps to better capture the motion cues between them. Then, the aligned features are fused to fill holes in the left image. To aggregate the multiscale feature maps dynamically, we propose a multiscale feature aggregation module based on an attention mechanism, of which the fusion module is designed as a symmetrical architecture to adaptively incorporate the complementary contextual correlations from different feature branches. In addition, a spatial attention module able to capture the correlations among pixels is introduced into our network to enhance the inpainting capacity and generate more refined details. To evaluate the effectiveness of our proposed method, many experiments are conducted on a stereo image dataset. The quantitative and qualitative results show that our method significantly outperforms the recent state-of-the-art image inpainting methods while running over 22 fps on a single NVIDIA RTX2080Ti GPU. • A new image inpainting method which utilizes additional image as constraints to assist the inpainting process is proposed. • The multi-scale feature alignment module based on deformable convolution can realize the feature alignment of different images. • Implicitly detect and restore damaged regions by exploiting attention maps. • Image textures and features are refined through a long-range spatial attention mechanism. [ABSTRACT FROM AUTHOR]

Autonomous robots must have the ability to build an accurate map of an unknown environment by fully covering it in an exploration task. Several exploration approaches combine a Simultaneous Localization and Mapping (SLAM) technique with a strategy to move the robot through the environment actively looking for loops to be closed. When closing a loop, the robot revisits a previously mapped area, which allows it to reduce the uncertainty about its pose. In this paper, we present a concise environment representation named Loop-Aware Exploration Graph (LAEG). The LAEG's nodes represent the essential information to the exploration process, such as the robot's position and the frontiers of two different kinds, while the LAEG's edges are the connections between these elements. Furthermore, the LAEG uses a specific type of edge to explicitly represent the predicted loops, facilitating the incorporation of this information into the exploration decision process. We also present an exploration approach that relies on the LAEG to make the decisions. Consequently, our approach maximizes the chances of closing a loop when choosing the next region to be explored, which is eased by the LAEG that represents the predicted loops as edges. The effectiveness of the proposed exploration approach was evaluated through experiments in five environments, comparing it with a greedy approach that only chases the most attractive unknown region and another one that makes the robot actively look for loop-closures. • The Loop-Aware Exploration Graph (LAEG) is a concise environment representation. • The LAEG represents only essential information for the exploration decision process. • The LAEG explicitly represents loop-closures hypotheses. • Loop-closures are essential to reduce the uncertainty about the robot's pose. [ABSTRACT FROM AUTHOR]

In an ageing society, the at-home use of Socially Assistive Robots (SARs) could provide remote monitoring of their users' well-being, together with physical and psychological support. However, private home environments are particularly challenging for SARs, due to their unstructured and dynamic nature which often contributes to robots' failures. For this reason, even though several prototypes of SARs for elderly care have been developed, their commercialisation and wide-spread at-home use are yet to be effective. In this paper, we analyse how including the end users' feedback impacts the SARs reliability and acceptance. To do so, we introduce a Monitoring and Logging System (MLS) for remote supervision, which increases the explainability of SAR-based systems deployed in older adults' apartments, while also allowing the exchange of feedback between caregivers, technicians, and older adults. We then present an extensive field study showing how long-term deployment of autonomous SARs can be accomplished by relying on such a feedback loop to address any potential issue. To this end, we provide the results obtained in a 130-week long study where autonomous SARs were deployed in the apartments of 10 older adults, with the aim of possibly serving and assisting future practitioners, with the knowledge collected from this extensive experimental campaign, to fill the gap that currently exists for the widespread adoption of SARs. • Socially assistive robots are promising tools for older adults independent living. • Private apartments are challenging environments, impacting the robustness of robots. • Results from a 130 weeks study in 10 older adults apartments using autonomous robots. • Remote supervision increased the acceptance of socially assistive robots. • Lessons learned from managing a long-term deployment, towards long-term autonomy. [ABSTRACT FROM AUTHOR]

This article presents a dataset collected from the subterranean (SubT) environment with a current state-of-the-art sensors required for autonomous navigation. The dataset includes sensor measurements collected with RGB, RGB-D, event-based and thermal cameras, 2D and 3D lidars, inertial measurement unit (IMU), and ultra wideband (UWB) positioning systems which are mounted on the mobile robot. The overall sensor setup will be referred further in the article as a data collection platform. The dataset contains synchronized raw data measurements from all the sensors in the robot operating system (ROS) message format and video feeds collected with action and 360 cameras. A detailed description of the sensors embedded into the data collection platform and a data collection process are introduced. The collected dataset is aimed for evaluating navigation, localization and mapping algorithms in SubT environments. This article is accompanied with the public release of all collected datasets from the SubT environment. Link: Dataset • The paper presents a dataset from the SubT environment with state-of-the-art sensors. • The dataset employs RGB, RGB-D, event-based and thermal cameras, lidars, IMU and UWB. • The dataset provides recorded data from all the sensors in ROS bag files. [ABSTRACT FROM AUTHOR]

This paper presents the dynamic modelling and linear matrix inequality (LMI) based controller design of a distributed fog computing framework for real-time robot vision applications. A mobile robot vision system acquires the images from an application environment such as a warehouse, where articles are stacked in numerous racks. We characterise the mobile robot vision data (MRVD) using frames per second (FPS) and the image resolution. From the MRVD, object detection is performed by an open-source deep learning (DL) platform for detecting and localising various objects. However, with higher FPS together with high-resolution images, the processing time by the DL algorithm increases significantly. This necessitates the deployment of a distributed computing platform with several computing nodes. In this work, we deploy a distributed fog computing environment (DFCE) for the real-time object detection in an application environment. The processing time required to handle the MRVD is called the service time. However, for efficient auto-scaling performance, the mathematical model of the DFCE, taking into consideration the characteristics of the MRVD is necessary. In this context, we envisage the application of control theory to build the dynamic model of the DFCE. A Linear Parameter Varying (LPV) model is proposed for the DFCE with the service time as the output, the number of fog nodes as the input, and the characteristics of MRVD as the time-varying parameters. At first, an LPV model for the DFCE is derived using system identification, and the model is validated using the real-time test data. The LPV model is converted to a Polytopic LPV (PLPV) model for LMI based controller design. Finally, we develop and validate a Linear Matrix Inequality (LMI) based LPV controller to meet the service time constraints for a given application environment. For localisation and trajectory tracking with obstacle avoidance in the application environment, the mobile robot implements an Extended Kalman Filter (EKF) based simultaneous localisation and mapping (SLAM) and a bug-based path planning algorithm respectively. Finally, we present, detailed controller validation results illustrating the mobile robot navigation together with the auto-scaling control for the fog computing platform to modulate the service time. [ABSTRACT FROM AUTHOR]

While current autonomous navigation systems allow robots to successfully drive themselves from one point to another in specific environments, they typically require extensive manual parameter re-tuning by human robotics experts in order to function in new environments. Furthermore, even for just one complex environment, a single set of fine-tuned parameters may not work well in different regions of that environment. These problems prohibit reliable mobile robot deployment by non-expert users. As a remedy, we propose Adaptive Planner Parameter Learning (appl), a machine learning framework that can leverage non-expert human interaction via several modalities – including teleoperated demonstrations, corrective interventions, and evaluative feedback – and also unsupervised reinforcement learning to learn a parameter policy that can dynamically adjust the parameters of classical navigation systems in response to changes in the environment. appl inherits safety and explainability from classical navigation systems while also enjoying the benefits of machine learning, i.e., the ability to adapt and improve from experience. We present a suite of individual appl methods and also a unifying cycle-of-learning scheme that combines all the proposed methods in a framework that can improve navigation performance through continual, iterative human interaction and simulation training. [ABSTRACT FROM AUTHOR]

This article addresses an unified solution to the trajectory tracking and path following problems for differential drive mobile robots (DDMR) considering a point of interest (PoI) with variable location relative to the vehicle. The mobile robot is modeled with an extended kinematic model avoiding typical singularities of this kind of vehicles, and allowing a straightforward definition of the corresponding inverse kinematics controller (IKC). This classical IKC fulfills the control objective with exponential error convergence but with the shortcoming of generating backward navigation when the PoI is located behind the DDMR, which is undesirable in some practical applications where the forward navigation must be preserved. This situation is theoretically analyzed, concluding that even though both forward and backward navigations correspond to equilibrium points of the closed loop, the stability of the forward navigation requires a PoI located in front of the DDMR, and the stability of the backward navigation requires a PoI located behind the DDMR. Finally, the article presents novel alternative controllers in order to always fulfill the motion objectives with stable forward navigation. Simulation results are presented to show the performance of the proposed controllers, and a real application of a robotic cane guiding its user is experimentally developed. • An unified motion control is defined for differential robots (DDMR). • The position of the point of interest (PoI) is considered variable. • Other controllers generate backward navigation if the PoI is behind the DDMR. • Our controller fulfills the motion objectives with stable forward navigation. • A real application of a robotic cane guiding its user is experimentally developed. [ABSTRACT FROM AUTHOR]

In this paper, we present an active visual SLAM approach for omnidirectional robots. The goal is to generate control commands that allow such a robot to simultaneously localize itself and map an unknown environment while maximizing the amount of information gained and consuming as low energy as possible. Leveraging the robot's independent translation and rotation control, we introduce a multi-layered approach for active V-SLAM. The top layer decides on informative goal locations and generates highly informative paths to them. The second and third layers actively re-plan and execute the path, exploiting the continuously updated map and local features information. Moreover, we introduce two utility formulations to account for the presence of obstacles in the field of view and the robot's location. Through rigorous simulations, real robot experiments, and comparisons with state-of-the-art methods, we demonstrate that our approach achieves similar coverage results with lesser overall map entropy. This is obtained while keeping the traversed distance up to 39% shorter than the other methods and without increasing the wheels' total rotation amount. Code and implementation details are provided as open-source, and all the generated data is available online for consultation. • Hierarchical Active SLAM framework. • Continuous active control of a robot for optimized heading directions. • Higher exploration amount per traveled meter. • Formulation of new utilities using the presence of obstacles in the field of view. [ABSTRACT FROM AUTHOR]

The important thing about mobile robotics is that its use satisfies rapid movement without collisions. Several methods have been designed for this purpose, but not all of them give the results expected by the user. Our contribution in this article is to allow the mobile robot to take advantage of two methods by the hybridization effect. These methods, implemented are ANFIS "Adaptive Neuro-Flou Inference System" and the PSO "Particle Swarm Optimization". Learning by the connectionist approach of the ANFIS controller is sufficient to allow the robot to reach its target but remains unsatisfactory. The contribution of the PSO algorithm allows the hybrid controller to optimize, the speed and therefore the positions of robot. The controller thus constituted gives better simulation results of the distances measured in pixels traveled by the hybrid method. [ABSTRACT FROM AUTHOR]

This work presents a novel stable controller for the person-following task that includes social considerations for a differential drive mobile robot equipped with an RGB-D camera and a laser range finder as main sensors. The proposed controller adapts its behavior based on the knowledge of both: a modified personal space distribution and human user velocity. Control objectives are focused hence on keeping the human user within the camera's field-of-view while the mobile robot follows it, with a socially acceptable motion through arbitrary paths. To show the good behavior of this proposal, simulation and real experimental results are included and discussed. The asymptotic stability of the overall system is proved through the Lyapunov theory. Also, in our proposal, three state-of-the-art algorithms were integrated with the controller. In particular, a new real-time multi-person skeletal tracking system is used to obtain the relative human–robot position, a text to speech algorithm is used to confirm the commands given by the human, and also, a SLAM algorithm is used to obtain the map of the environment while the main task is being performed. Additionally, a hand gesture recognition module is included to interact with the mobile robot. This way, the robot is allowed to navigate with a socially-aware behavior in environments shared with humans. Finally, subjective and objective metrics are used as a validation method for human perception about the achieved robot motion. • The kinematic controller, derived from a new concept of "extended kinematic model". • The variable gains which, depending on the relative velocities, automatically adjusts the controller behavior. • The use of proxemics to give a comfort sensation to the user. • The stability proof with the reporting of real robot's experimental results. • Subjective and objective metric assessments from the users to conclude that our social robot navigation in the person-following task is accepted. [ABSTRACT FROM AUTHOR]

This paper presents a systematic study about the applicability of color/texture descriptors in a global localization system based on planar surface segments. Two comprehensive experiments regarding matching of planar surface segments and robot pose hypothesis evaluation were conducted. The experiments show that using color/texture descriptors to prune potential surface pairs in the initial correspondence phase and to provide additional information in the hypothesis evaluation phase of a feature-based localization scheme can result in considerable speedup of the localization process and help distinguish between geometrically similar places. An experimental benchmark which enables researchers to evaluate the performance of color and texture descriptors in the context of mobile robot localization based on planar surface segments is presented. [ABSTRACT FROM AUTHOR]

This work offers the solution at the control feed-back level of the accurate positioning in a finite time of the end-effector whose mobile manipulator is subject to control and complex state constraints. We propose new forms of various terminal sliding modes (TSM’s) which result from the access to kinematic redundancy of the non-holonomic mechanical system. In order to incorporate control and state inequality constraints into trajectory generation law, both a suitably defined extended task error is introduced and exterior penalty function approach is utilized. In addition, to incorporate holonomic singularity avoidance condition, collision avoidance constraints for the whole mobile manipulator and its final velocity, a suitably defined projection term onto the null space of the extended Jacobian matrix has been introduced. Control limits are maintained by suitable choice of trajectory generator gains. The numerical simulation results carried out for a mobile manipulator consisting of a nonholonomic differentially steered wheeled mobile platform and a holonomic manipulator of two revolute kinematic pairs, operating both in a two-dimensional unconstrained work space and work space including the obstacles, illustrate performance of the proposed trajectory generators. [ABSTRACT FROM AUTHOR]

This paper considers the problem of robust adaptive trajectory tracking control for a wheeled mobile robot (WMR). First, the trajectory tracking of the WMR is converted to a problem of the stabilization of a double integral system. Next, a continuous finite-time control method is employed to design a tracking controller. Then, a disturbance observer and an adaptive compensator are designed to cooperate with the tracking controller for dealing with system uncertainties of the WMR. Finally, a switching adaptive law is presented in combination with the boundary layer approach to attenuate the chattering in the adaptive compensator. As a result, the control system yields the ultimate boundedness of both the tracking error and the adaptive gain. Simulation results demonstrate the validity of the new method. [ABSTRACT FROM AUTHOR]

Mobile robot platforms capable of operating safely and accurately in dynamic environments can have a multitude of applications, ranging from simple delivery tasks to advanced assembly operations. These abilities rely heavily on a robust navigation stack, which requires stable and accurate pose estimations within the environment. To solve this problem, a modular localization system suitable for a wide range of mobile robot platforms was developed. By using LIDAR/RGB-D data, the proposed system is capable of achieving 1–2 cm in translation error and 1°–3° degrees in rotation error while requiring only 5–35 ms of processing time (in 3 and 6 DoF respectively). The system was tested in three robot platforms and in several environments with different sensor configurations. It demonstrated high accuracy while performing pose tracking with point cloud registration algorithms and high reliability when estimating the initial pose using feature matching techniques. The system can also build a map of the environment with surface reconstruction and incrementally update it with either the full field of view of the sensor data or only the unknown sections, which allows to reduce the mapping processing time and also gives the possibility to update a CAD model of the environment without degrading the detail of known static areas due to sensor noise. [ABSTRACT FROM AUTHOR]

Deformable slope is a type of terrain that wheeled mobile robots (WMRs) and ground unmanned vehicles (GUVs) may have to traverse to accomplish their mission tasks. However, the associated terramechanics for wheels with arbitrary posture is rarely studied. In this paper, based on wheel attitude, dynamics of the wheel–terrain interaction for a rigid wheel on deformable slope is investigated. Through introducing the angular geometry of wheel attitude into terramechanics theory, a generalized dynamic model is developed, involving two inclination angles of slope and three attitude angles of wheel steering axis. Two representative cases are studied: the wheel runs straight forward and perpendicular to the slope, and the wheel is in a steering maneuver with an inclined steering axis. A generalized kinematic model for wheel–terrain contact point and wheel center is also provided, which analytically explicates that trajectory of wheel motion is coupled with wheel attitude while driven by angular rates. The proposed attitude-based models are valid for arbitrary wheel–terrain geometry and can lead to control purpose directly. Effectiveness of the models is confirmed by simulating the influences from attitude to wheel mechanics and motion. [ABSTRACT FROM AUTHOR]

The use of 3D data in mobile robotics applications provides valuable information about the robot’s environment. However usually the huge amount of 3D information is difficult to manage due to the fact that the robot storage system and computing capabilities are insufficient. Therefore, a data compression method is necessary to store and process this information while preserving as much information as possible. A few methods have been proposed to compress 3D information. Nevertheless, there does not exist a consistent public benchmark for comparing the results (compression level, distance reconstructed error, etc.) obtained with different methods. In this paper, we propose a dataset composed of a set of 3D point clouds with different structure and texture variability to evaluate the results obtained from 3D data compression methods. We also provide useful tools for comparing compression methods, using as a baseline the results obtained by existing relevant compression methods. [ABSTRACT FROM AUTHOR]

A path planning in the space of multinominals integrated into a model predictive control mechanism for driving formations of autonomous mobile robots is presented in this paper. The proposed approach is designed to stabilize the formations in desired shapes, and to navigate the group into a final position in a partly known environment with dynamic obstacles. In addition, the system provides inter-vehicle coordination and collision avoidance in the event of failure of a team member. The method is aimed at reaching the final position of the formation in the desired shape, but it enables to change temporarily this shape if it is enforced by the environment (in narrow corridors, on response to an impending collision with obstacles and faulty team members, etc.). This autonomous emergent behaviour increases the robustness of the system and its usability. It enables a proper compromise to be found between the formation driving requirement and the effort to fulfil the mission objective, i.e., to move the group from the current state into the required position. In this paper, the convergence of the method and the requirements for stability are shown on the basis of the results of the Lyapunov theorems of stability. These theoretical achievements imply constraints on the applicability of the system, which are verified in numerical simulations and in various tests with real autonomous robots. The performances of the entire system and of independent sub-systems in various formation driving scenarios are also shown in these tests. [ABSTRACT FROM AUTHOR]

Autonomous mobile robots (AMRs) interacting with an a priori distributed wireless sensor network (WSN) in a region can address the three-tier challenge of navigating in unknown environments: (i) identifying target locations, (ii) planning paths to the targets, and (iii) efficiently executing the navigation paths to the targets. This paper presents low-complexity algorithms to address the second-tier and third-tier challenges, i.e., efficiently planning and executing paths to target locations. These novel approaches use only the information inherent in WSNs, i.e., received signal strength (RSS). The objective is to have the AMR navigate to a target location by: (i) producing an RSS-based artificial magnitude distribution in the navigation region, (ii) using particle filtering based bearing estimation for orientation information, and (iii) using interpolated pseudogradient for efficient path planning and navigation. Here, the AMR does not require: (i) the global location information for itself or the WSN, (ii) a priori information of the direction of a target location, or (iii) sophisticated ranging equipment for prior mapping. The AMR relies only on local, neighborhood information and low-cost wireless directional antennas for navigation. Real-world and simulation experiments, using a variety of node-densities, demonstrate the effectiveness of the proposed schemes. The low-cost, low-complexity advantages of the WSN–AMR interactive navigation provide for efficient map-less and ranging-less navigation methods. [ABSTRACT FROM AUTHOR]