Abstract: Model predictive control (MPC) is an effective method for controlling robotic systems, particularly autonomous aerial vehicles such as quadcopters. However, application of MPC can be computationally demanding, and typically requires estimating the state of the system, which can be challenging in complex, unstructured environments. Reinforcement learning can in principle forego the need for explicit state estimation and acquire a policy that directly maps sensor readings to actions, but is difficult to apply to unstable systems that are liable to fail catastrophically during training before an effective policy has been found. We propose to combine MPC with reinforcement learning in the framework of guided policy search, where MPC is used to generate data at training time, under full state observations provided by an instrumented training environment. This data is used to train a deep neural network policy, which is allowed to access only the raw observations from the vehicle's onboard sensors. After training, the neural network policy can successfully control the robot without knowledge of the full state, and at a fraction of the computational cost of MPC. We evaluate our method by learning obstacle avoidance policies for a simulated quadrotor, using simulated onboard sensors and no explicit state estimation at test time.

Abstract: Steer-by-wire technology enables vehicle safety systems to share control with a driver through augmentation of the driver's steering commands. Advances in sensing technologies empower these systems further with real-time information about the surrounding environment. Leveraging these advancements in vehicle actuation and sensing, the authors present a shared control framework for obstacle avoidance and stability control using two safe driving envelopes. One of these envelopes is defined by the vehicle handling limits, whereas the other is defined by spatial limitations imposed by lane boundaries and obstacles. A model predictive control (MPC) scheme determines at each time step if the current driver command allows for a safe vehicle trajectory within these two envelopes, intervening only when such a trajectory does not exist. In this way, the controller shares control with the driver in a minimally invasive manner while avoiding obstacles and preventing loss of control. The optimal control problem underlying the controller is inherently nonconvex but is solved as a set of convex problems allowing for reliable real-time implementation. This approach is validated on an experimental vehicle working with human drivers to negotiate obstacles in a low friction environment.

Abstract: A key skill for mobile robots is the ability to navigate efficiently through their environment. In the case of social or assistive robots, this involves navigating through human crowds. Typical performance criteria, such as reaching the goal using the shortest path, are not appropriate in such environments, where it is more important for the robot to move in a socially adaptive manner such as respecting comfort zones of the pedestrians. We propose a framework for socially adaptive path planning in dynamic environments, by generating human-like path trajectory. Our framework consists of three modules: a feature extraction module, inverse reinforcement learning (IRL) module, and a path planning module. The feature extraction module extracts features necessary to characterize the state information, such as density and velocity of surrounding obstacles, from a RGB-depth sensor. The inverse reinforcement learning module uses a set of demonstration trajectories generated by an expert to learn the expert’s behaviour when faced with different state features, and represent it as a cost function that respects social variables. Finally, the planning module integrates a three-layer architecture, where a global path is optimized according to a classical shortest-path objective using a global map known a priori, a local path is planned over a shorter distance using the features extracted from a RGB-D sensor and the cost function inferred from IRL module, and a low-level system handles avoidance of immediate obstacles. We evaluate our approach by deploying it on a real robotic wheelchair platform in various scenarios, and comparing the robot trajectories to human trajectories.

Abstract: Formation control is a cooperative control concept in which multiple autonomous underwater mobile robots are deployed for a group motion and/or control mission. This paper presents a brief review on various cooperative search and formation control strategies for multiple autonomous underwater vehicles (AUV) based on literature reported till date. Various cooperative and formation control schemes for collecting huge amount of data based on formation regulation control and formation tracking control are discussed. To address the challenge of detecting AUV failure in the fleet, communication issues, collision and obstacle avoidance are also taken into attention. Stability analysis of the feasible formation is also presented. This paper may be intended to serve as a convenient reference for the further research on formation control of multiple underwater mobile robots.

Abstract: We propose a new path planning algorithm based on the use of Q-learning and artificial neural networks.We analyze and model in VR the mobile robot PowerBot.We implement and test the proposed algorithm in both VR and real workspaces.The solution converges to collision-free trajectories in dynamic environments. This study proposes a new approach for solving the problem of autonomous movement of robots in environments that contain both static and dynamic obstacles. The purpose of this research is to provide mobile robots a collision-free trajectory within an uncertain workspace which contains both stationary and moving entities. The developed solution uses Q-learning and a neural network planner to solve path planning problems. The algorithm presented proves to be effective in navigation scenarios where global information is available. The speed of the robot can be set prior to the computation of the trajectory, which provides a great advantage in time-constrained applications. The solution is deployed in both Virtual Reality (VR) for easier visualization and safer testing activities, and on a real mobile robot for experimental validation. The algorithm is compared with Powerbot's ARNL proprietary navigation algorithm. Results show that the proposed solution has a good conversion rate computed at a satisfying speed.

Abstract: Miniature Micro Aerial Vehicles (MAV) are very suitable for flying in indoor environments, but autonomous navigation is challenging due to their strict hardware limitations. This paper presents a highly efficient computer vision algorithm called Edge-FS for the determination of velocity and depth. It runs at 20 Hz on a 4 g stereo camera with an embedded STM32F4 microprocessor (168 MHz, 192 kB) and uses feature histograms to calculate optical flow and stereo disparity. The stereo-based distance estimates are used to scale the optical flow in order to retrieve the drone's velocity. The velocity and depth measurements are used for fully autonomous flight of a 40 g pocket drone only relying on on-board sensors. The method allows the MAV to control its velocity and avoid obstacles.

Abstract: This paper introduces a reinforcement learning method for exploring a corridor environment with the depth information from an RGB-D sensor only. The robot controller achieves obstacle avoidance ability by pre-training of feature maps using the depth information. The system is based on the recent Deep Q-Network (DQN) framework where a convolution neural network structure was adopted in the Q-value estimation of the Q-learning method. We separate the DQN into a supervised deep learning structure and a Q-learning network. The experiments of a Turtlebot in the Gazebo simulation environment show the robustness to different kinds of corridor environments. All of the experiments use the same pre-training deep learning structure. Note that the robot is traveling in environments which are different from the pre-training environment. It is the first time that raw sensor information is used to build such an exploring strategy for robotics by reinforcement learning.

Abstract: The shortest/optimal path planning is essential for efficient operation of autonomous vehicles. In this article, a new nature-inspired meta-heuristic algorithm has been applied for mobile robot path planning in an unknown or partially known environment populated by a variety of static obstacles. This meta-heuristic algorithm is based on the levy flight behaviour and brood parasitic behaviour of cuckoos. A new objective function has been formulated between the robots and the target and obstacles, which satisfied the conditions of obstacle avoidance and target-seeking behaviour of robots present in the terrain. Depending upon the objective function value of each nest (cuckoo) in the swarm, the robot avoids obstacles and proceeds towards the target. The smooth optimal trajectory is framed with this algorithm when the robot reaches its goal. Some simulation and experimental results are presented at the end of the paper to show the effectiveness of the proposed navigational controller.

Abstract: This brief presents a novel control scheme for some problems on tracking and obstacle avoidance of a wheeled mobile robot with nonholonomic constraint. An extended state observer is introduced to estimate the unknown disturbances and velocity information of the wheeled mobile robot. A nonlinear controller is designed to achieve tracking target and obstacle avoidance in complex environments. Note that tracking errors converge to a residual set outside the obstacle detection region. Moreover, the obstacle avoidance is also guaranteed inside the obstacle detection region. Simulation results are given to verify the effectiveness and robustness of the proposed design scheme.

Abstract: Accurate closed-loop control of continuum manipulators requires integration of both models that describe their motion and methods to evaluate manipulator shape. This work presents a model that approximates the continuous shape of a continuum manipulator by a serial chain of rigid links, connected by flexible rotational joints. This rigid-link model permits a description of manipulator shape under different loading conditions. A kinematic controller, based on the manipulator Jacobian of the proposed rigid-link model, is implemented and realizes trajectory tracking, while using the kinematic redundancy of the manipulator to perform a secondary task of avoiding obstacles. The controller is evaluated on an experimental testbed, consisting of a planar tendon-driven continuum manipulator with two bending segments. Fiber Bragg grating (FBG) sensors are used to reconstruct 3-D manipulator shape, and is used as feedback for closed-loop control of the manipulator. Manipulator steering is evaluated for two cases: the first case involving steering around a static obstacle and the second case involving steering along a straight path while avoiding a moving obstacle. Mean trajectory tracking errors are 0.24 and 0.09 mm with maximum errors of 1.37 and 0.52 mm for the first and second cases, respectively. Finally, we demonstrate the possibility of FBG sensors to measure interaction forces, while simultaneously using them for shape sensing.

Abstract: This paper describes the design and the implementation of a trajectory tracking controller using fuzzy logic for mobile robot to navigate in indoor environments. Most of the previous works used two independent controllers for navigation and avoiding obstacles. The main contribution of the paper can be summarized in the fact that we use only one fuzzy controller for navigation and obstacle avoidance. The used mobile robot is equipped with DC motor, nine infrared range (IR) sensors to measure the distance to obstacles, and two optical encoders to provide the actual position and speeds. To evaluate the performances of the intelligent navigation algorithms, different trajectories are used and simulated using MATLAB software and SIMIAM navigation platform. Simulation results show the performances of the intelligent navigation algorithms in terms of simulation times and travelled path.

Abstract: This paper describes a vision-based obstacle detection and navigation system for use as part of a robotic solution for the sustainable intensification of broad-acre agriculture. To be cost-effective, the robotics solution must be competitive with current human-driven farm machinery. Significant costs are in high-end localization and obstacle detection sensors. Our system demonstrates a combination of an inexpensive global positioning system and inertial navigation system with vision for localization and a single stereo vision system for obstacle detection. The paper describes the design of the robot, including detailed descriptions of three key parts of the system: novelty-based obstacle detection, visually-aided guidance, and a navigation system that generates collision-free kinematically feasible paths. The robot has seen extensive testing over numerous weeks of field trials during the day and night. The results in this paper pertain to one particular 3 h nighttime experiment in which the robot performed a coverage task and avoided obstacles. Additional results during the day demonstrate that the robot is able to continue operating during 5 min GPS outages by visually following crop rows.

Abstract: This paper considers the problem of optimal trajectory generation for autonomous driving under both continuous and logical constraints. Classical approaches based on continuous optimization formulate the trajectory generation problem as a nonlinear program, in which vehicle dynamics and obstacle avoidance requirements are enforced as nonlinear equality and inequality constraints. In general, gradient-based optimization methods are then used to find the optimal trajectory. However, these methods are ill-suited for logical constraints such as those raised by traffic rules, presence of obstacles and, more generally, to the existence of multiple maneuver variants. We propose a new formulation of the trajectory planning problem as a Mixed-Integer Quadratic Program. This formulation can be solved efficiently using widely available solvers, and the resulting trajectory is guaranteed to be globally optimal. We apply our framework to several scenarios that are still widely considered as challenging for autonomous driving, such as obstacle avoidance with multiple maneuver choices, overtaking with oncoming traffic or optimal lane-change decision making. Simulation results demonstrate the effectiveness of our approach and its real-time applicability.

Abstract: Exploration in an unknown environment is an elemental application for mobile robots. In this paper, we outlined a reinforcement learning method aiming for solving the exploration problem in a corridor environment. The learning model took the depth image from an RGB-D sensor as the only input. The feature representation of the depth image was extracted through a pre-trained convolutional-neural-networks model. Based on the recent success of deep Q-network on artificial intelligence, the robot controller achieved the exploration and obstacle avoidance abilities in several different simulated environments. It is the first time that the reinforcement learning is used to build an exploration strategy for mobile robots through raw sensor information.

Abstract: Real-time 3D perception is critical for localisation, mapping, path planning and obstacle avoidance for mobile robots and autonomous vehicles. For outdoor operation in real-world environments, 3D perception is often provided by sparse 3D LIDAR scanners, which provide accurate but low-density depth maps, and dense stereo approaches, which require significant computational resources for accurate results. Here, taking advantage of the complementary error characteristics of LIDAR range sensing and dense stereo, we present a probabilistic method for fusing sparse 3D LIDAR data with stereo images to provide accurate dense depth maps and uncertainty estimates in real-time. We evaluate the method on data collected from a small urban autonomous vehicle and the KITTI dataset, providing accuracy results competitive with state-of-the-art stereo approaches and credible uncertainty estimates that do not misrepresent the true errors, and demonstrate real-time operation on a range of low-power GPU systems.

Abstract: Inverse kinematics algorithms are commonly used in robotic systems to transform tasks to joint references, and several methods exist to ensure the achievement of several tasks simultaneously. The multiple task-priority inverse kinematics framework allows tasks to be considered in a prioritized order by projecting task velocities through the nullspaces of higher priority tasks. This paper extends this framework to handle setbased tasks, i.e. tasks with a range of valid values, in addition to equality tasks, which have a specific desired value. Examples of set-based tasks are joint limit and obstacle avoidance. The proposed method is proven to ensure asymptotic convergence of the equality task errors and the satisfaction of all high-priority set-based tasks. The practical implementation of the proposed algorithm is discussed, and experimental results are presented where a number of both set-based and equality tasks have been implemented on a 6 degree of freedom UR5 which is an industrial robotic arm from Universal Robots. The experiments validate the theoretical results and confirm the effectiveness of the proposed approach.

Abstract: In robotics, Ultrasonic sensors and Infrared sensors are commonly used for distance measurement. These low-cost sensors fundamentally address majority of problems related to the obstacle detection and obstacle avoidance. In this paper, the performance comparison of ultrasonic and infrared measurement techniques across obstacles of different types of materials presented. The Vehicle model integrated with the sensors, moving with constant velocity towards different types of obstacles for capturing the distance parameter. Based on the data acquired from the sensors, correlation analysis of the measured distance with actual distance performed. This analysis will be very much useful, to select the right sensor - Ultrasonic sensor / Infrared sensor or a combination of both sensors, while developing the algorithm for addressing obstacle detection problems. The detection range and inherent properties of sensors (reflection/ absorption etc.) also were tested in this experiment.

Abstract: The requirement to increase inspection speeds for non-destructive testing (NDT) of composite aerospace parts is common to many manufacturers. The prevalence of complex curved surfaces in the industry provides motivation for the use of 6 axis robots in these inspections. The purpose of this paper is to present work undertaken for the development of a KUKA robot manipulator based automated NDT system. A new software solution is presented that enables flexible trajectory planning to be accomplished for the inspection of complex curved surfaces often encountered in engineering production. The techniques and issues associated with conventional manual inspection techniques and automated systems for the inspection of large complex surfaces were reviewed. This approach has directly influenced the development of a MATLAB toolbox targeted to NDT automation, capable of complex path planning, obstacle avoidance, and external synchronization between robots and associated external NDT systems. This paper highlights the advantages of this software over conventional off-line-programming approaches when applied to NDT measurements. An experimental validation of path trajectory generation, on a large and curved composite aerofoil component, is presented. Comparative metrology experiments were undertaken to evaluate the real path accuracy of the toolbox when inspecting a curved 0.5 m2 and a 1.6 m2 surface using a KUKA KR16 L6-2 robot. The results have shown that the deviation of the distance between the commanded TCPs and the feedback positions were within 2.7 mm. The variance of the standoff between the probe and the scanned surfaces was smaller than the variance obtainable via commercial path-planning software. Tool paths were generated directly on the triangular mesh imported from the CAD models of the inspected components without need for an approximating analytical surface. By implementing full external control of the robotic hardware, it has been possible to synchronise the NDT data collection with positions at all points along the path, and our approach allows for the future development of additional functionality that is specific to NDT inspection problems. For the current NDT application, the deviations from CAD design and the requirements for both coarse and fine inspections, dependent on measured NDT data, demand flexibility in path planning beyond what is currently available from existing off-line robot programming software.

Abstract: Automatic obstacle detection and avoidance is a key component for the success of micro-aerial vehicles (MAVs) in the future. As the payload of MAVs is highly constrained, cameras are attractive sensors because they are both lightweight and provide rich information about the environment. In this paper, we present an approach that allows a quadrotor with a single monocular camera to locally generate collision-free waypoints. We acquire a small set of images while the quadrotor is hovering from which we compute a dense depth map. Based on this depth map, we render a 2D scan and generate a suitable waypoint for navigation. In our experiments, we found that the pose variation during hovering is already sufficient to obtain suitable depth maps. The computation takes less than one second which renders our approach applicable for obstacle avoidance in real-time. We demonstrate the validity of our approach in challenging environments where we navigate a Parrot Ardrone quadrotor successfully through narrow passages including doors, boxes, and people.

Abstract: This paper addresses the formation control problem of multiple Euler-Lagrange systems with model uncertainties in the environment containing obstacles. Utilizing the null-space-based (NSB) behavioral control architecture, the proposed problem can be decomposed into elementary missions (behaviors) with different priorities and implemented by each individual system. A class of novel coordination control algorithms is constructed and utilized to achieve accurate formation task while avoiding obstacles and guaranteeing the model uncertainty rejection objective. By using sliding mode control and Lyapunov theory, the formation performance in closed-loop multi-agent systems is proven achievable if the state-dependent gain of the obstacle avoidance mission is appropriately designed. Finally, simulation examples demonstrate the effectiveness of the algorithms.

Abstract: In this part, we present the three main contribution of the paper:The Fuzzy-PI for linear velocityThe Fuzzy-PI for angular speedObstace avoidance controller based on Mamdani-fuzzy controller which allows the omnidirectionnal mobile robot to avoid fixed obstacles in an unknown environment.Display Omitted Most of the previous works are focused on only the estimation of the final navigation position of the mobile robot.The omnidirectional robot has to reach the final desired position with a predefined final angle.The optimization of the travelled distance using an appropriate FLC.Simulation and experimental tests are performed for one, two and three obstacles to evaluate the real performances of the developed algorithms. This paper presents a new design approach for an intelligent navigation algorithm for omnidirectional mobile robots. Unlike the previous works dealing with the navigation of omnidirectional robots that are focused on only the estimation of the final position, the main contribution of the present study is summed up in the fact that the robot has to reach the final desired position with a predefined final steering angle. This latter improvement is a part of researches carried out on Intelligent Transport Systems (ITS). Taking into consideration the drawbacks of proportional integral (PI) control when applied to omnidirectional robot navigation, we develop an approach to design a fuzzy logic PI controller (Fuzzy-PI).Preliminary simulation and experimental results using the Fuzzy-PI controller have shown limitations as the robot performed a larger path when the desired final angle increased. Thus, a deepen study have concluded that the Fuzzy-PI system cannot control at the same time the linear and angular velocities. To overcome these drawbacks, we propose to replace the previous intelligent navigation system by two independent controllers. The designed Fuzzy-PI controllers can adjust their parameters (KP, KI) to reduce the error caused by the dynamic changes and navigation challenges of omnidirectional robot. A navigation algorithm cannot be efficient without obstacle avoidance system. To achieve this goal, we have developed a third fuzzy controller to fulfill this task.To evaluate the real performances of fuzzy controllers for navigation and obstacles avoidance, simulation and experimental tests are performed for one, two and three obstacles. The obtained results confirm that the omnidirectional robot could navigate in an unstructured and unknown obstacles environment with better performances and efficiency.

Abstract: In recent years, the availability of low-cost and autonomous unmanned aerial vehicles (UAVs) results in the use of them for different types of military and commercial applications. The crucial part of the autonomous UAVs is their online or offline path planning algorithms. In the literature, there are many types of solutions, which use evolutionary and/or swarm intelligence approaches. Ant colony optimization is one of the mostly used algorithms, which has been applied to solve different type of path planning problems. Mainly, most of these studies have focused on a single colony ant colony optimization (ACO), which can find better solutions in fewer computation times. However, it is able to converge to a sub-optimal solution in the planning process. One approach to avoid the premature convergence is the use of Multi-Colony ACO, in which a number of ant colonies try to find an optimal solution cooperatively by exchanging their valuable information with each other. In this paper, it is aimed to implement an obstacle avoidance UAV path planning by using Multi-Colony ACO algorithm. We experimentally investigate the use of Multi-Colony ACO approach results from an effective path planning for UAVs with a comparison to a single colony ACO approach.

Abstract: In this letter, we investigate the semitied Gaussian mixture models for robust learning and adaptation of robot manipulation tasks. We make use of the spatial and temporal correlation in the data by tying the covariance matrices of the mixture model with common synergistic directions/basis vectors, instead of estimating full covariance matrices for each cluster in the mixture. This allows the reuse of the discovered synergies in different parts of the task having similar coordination patterns. We extend the approach to task-parameterized and hidden semi-Markov models for autonomous adaptation to changing environmental situations. The planned movement sequence from the model is smoothly followed with a finite horizon linear quadratic tracking controller. Experiments to encode whole body motion data in simulation, followed by valve opening and pick-and-place via obstacle avoidance tasks with the Baxter robot, show improvement over standard Gaussian mixture models with much less parameters and better generalization ability.

Abstract: This paper investigates the level of model fidelity needed in order for a model predictive control (MPC)-based obstacle avoidance algorithm to be able to safely and quickly avoid obstacles even whe...