Dynamic Obstacle Avoidance Research Articles

Improved DQN for Dynamic Obstacle Avoidance and Ship Path Planning

The avoidance of collisions among ships requires addressing various factors such as perception, decision-making, and control. These factors pose many challenges for autonomous collision avoidance. Traditional collision avoidance methods have encountered significant difficulties when used in autonomous collision avoidance. They are challenged to cope with the changing environment and harsh motion constraints. In the actual navigation of ships, it is necessary to carry out decision-making and control under the constraints of ship manipulation and risk. From the implementation process perspective, it is a typical sequential anthropomorphic decision-making problem. In order to solve the sequential decision problem, this paper improves DQN by setting a priority for sample collection and adopting non-uniform sampling, and it is applied to realize the intelligent collision avoidance of ships. It also verifies the performance of the algorithm in the simulation environment.

Cooperative strategy for pursuit-evasion problem in the presence of static and dynamic obstacles

In complex and dynamic marine environments with multiple obstacles, efficient and flexible pursuit is a challenging task. This paper proposes a self-organizing cooperative pursuit strategy. First, the Apollonius circle is introduced as the final formation of the encirclement pattern, and the trajectory of the target is predicted on the basis of the least-squares method. To improve the efficiency of collision avoidance against static irregular obstacles and obtain a more reasonable collision avoidance path, we design a discretization method to describe the boundary of obstacles and combine it with the artificial potential field (APF) method. Furthermore, dynamic obstacle avoidance is considered to extend the applicability of the algorithm to a variety of obstacle environments. Compared with the common greedy algorithm, the simulation results show that in the obstacle free and static obstacle environment, the self-organizing cooperative pursuit algorithm can shorten the pursuit time and improve the movement consistency of the USVs swarm, and the obstacle avoidance route is smoother. In complex environments containing static obstacles and dynamic ships, the self-organizing cooperative pursuit algorithm can also pursuit escape targets effectively.

Trajectory Planning and Tracking for Four-Wheel Independent Drive Intelligent Vehicle Based on Model Predictive Control

<div class="section abstract"><div class="htmlview paragraph">This paper proposes a dynamic obstacle avoidance system to help autonomous vehicles drive on high-speed structured roads. The system is mainly composed of trajectory planning and tracking controllers. The potential field (PF) model is introduced to establish a three-dimensional potential field for structured roads and obstacle vehicles. The trajectory planning problem that considers the vehicle’s and tires’ dynamics constraints is transformed into an optimization problem with muti-constraints by combining the model predictive control (MPC) algorithms. The trajectory tracking controller used in this paper is based on the 7 degrees of freedom (DOF) vehicle model and the UniTire tire model, which was discussed in detail in previous work [<span class="xref">25</span>, <span class="xref">26</span>]. The controller maintains good trajectory tracking performance even under extreme driving conditions, such as roads with poor adhesion conditions, where the car’s tires enter the nonlinear region easily. The innovation of this paper lies in introducing a high-fidelity vehicle model and a muti-conditions UniTire tire model with high precision obtained after parameter identification through tire test data. In addition, the nonlinear relationship between tire lateral force and slip angle is expressed as a linear function. This method updates the equivalent cornering stiffness in the linear function by solving the slope of the secant of the tire cornering characteristic curve at the current moment online, which solves the problem of high computational complexity caused by applying complex tire models in the control system. The co-simulation results of Simulink and CarSim show that the designed system has good dynamic obstacle avoidance and driving stability performance under driving conditions with high speed and poor road adhesion.</div></div>

A Double-Layer Model Predictive Control Approach for Collision-Free Lane Tracking of On-Road Autonomous Vehicles

This paper proposes a double-layer model predictive control (MPC) algorithm for the integrated path planning and trajectory tracking of autonomous vehicles on roads. The upper module is responsible for generating collision-free lane trajectories, while the lower module is responsible for tracking this trajectory. A simplified vehicle model based on the friction cone is proposed to reduce the computation time for trajectory planning in the upper layer module. To achieve dynamic and accurate collision avoidance, a polygonal distance-based dynamic obstacle avoidance method is proposed. A vertical load calculation method for the tires is introduced to design the anti-rollover constraint in the lower layer module. Numerical simulations, with static and dynamic obstacle scenarios, are conducted on the MATLAB platform and compared with two state-of-the-art MPC algorithms. The results demonstrate that the proposed algorithm outperforms the other two algorithms regarding computation time and collision avoidance efficiency.

Path Planning for Obstacle Avoidance of Robot Arm Based on Improved Potential Field Method.

In medical and surgical scenarios, the trajectory planning of a collaborative robot arm is a difficult problem. The artificial potential field (APF) algorithm is a classic method for robot trajectory planning, which has the characteristics of good real-time performance and low computing consumption. There are many variants of the APF algorithm, among which the most widely used variants is the velocity potential field (VPF) algorithm. However, the traditional VPF algorithm has inherent defects and problems, such as easily falling into local minimum, being unable to reach the target, poor dynamic obstacle avoidance ability, and safety and efficiency problems. Therefore, this work presents the improved velocity potential field (IVPF) algorithm, which considers direction factors, obstacle velocity factor, and tangential velocity. When encountering dynamic obstacles, the IVPF algorithm can avoid obstacles better to ensure the safety of both the human and robot arm. The IVPF algorithm also does not easily fall into a local problem when encountering different obstacles. The experiments informed the RRT* algorithm, VPF algorithm, and IVPF algorithm for comparison. Compared with the informed RRT* and VPF algorithm, the result of experiments indicate that the performances of the IVPF algorithm have significant improvements when dealing with different obstacles. The main aim of this paper is to provide a safe and efficient path planning algorithm for the robot arm in the medical field. The proposed algorithm can ensure the safety of both the human and the robot arm when the medical and surgical robot arm is working, and enables the robot arm to cope with emergencies and perform tasks better. The application of the proposed algorithm could make the collaborative robots work in a flexible and safe condition, which could open up new opportunities for the future development of medical and surgical scenarios.

Path Following and Avoiding Obstacle for Mobile Robot Under Dynamic Environments Using Reinforcement Learning

Obstacle avoidance for mobile robot to reach the desired target from a start location is one of the most interesting research topics. However, until now, few works discuss about working of mobile robot in the dynamic and continuously changing environment. So, this issue is still the research challenge for mobile robots. Traditional algorithm for obstacle avoidance in the dynamic, complex environment had many drawbacks. As known that Q-learning, the type of reinforcement learning, has been successfully applied in computer games. However, it is still rarely used in real world applications. This research presents an effectively method for real time dynamic obstacle avoidance based on Q-learning in the real world by using three-wheeled mobile robot. The position of obstacles including many static and dynamic obstacles and the mobile robot are recognized by fixed camera installed above the working space. The input for the robot is the 2D data from the camera. The output is an action for the robot (velocities, linear and angular parameters). Firstly, the simulation is performed for Q-learning algorithm then based on trained data, The Q-table value is implemented to the real mobile robot to perform the task in the real scene. The results are compared with intelligent control method for both static and dynamic obstacles cases. Through implement experiments, the results show that, after training in dynamic environments and testing in a new environment, the mobile robot is able to reach the target position successfully and have better performance comparing with fuzzy controller.

An online collision-free trajectory generation algorithm for human–robot collaboration

The premise of human–robot collaboration is that robots have adaptive trajectory planning strategies in hybrid work cell. The aim of this paper is to propose a new online collision avoidance trajectory planning algorithm for moderate dynamic environments to insure human safety when sharing collaborative tasks. The algorithm contains two parts: trajectory generation and local optimization. Firstly, based on empirical Dirichlet Process Gaussian Mixture Model (DPGMM) distribution learning, a neural network trajectory planner called Collaborative Waypoint Planning network (CWP-net) is proposed to generate all key waypoints required for dynamic obstacle avoidance in joint space according to environmental inputs. These points are used to generate quintic spline smooth motion trajectories with velocity and acceleration constraints. Secondly, we present an improved Stochastic Trajectory Optimization for Motion Planning (STOMP) algorithm which locally optimizes the generated trajectories of CWP-net by constraining the trajectory optimization range and direction through the DPGMM model. Simulations and real experiments from an industrial use case of human–robot collaboration in the field of aircraft assembly testing show that the proposed algorithm can smoothly adjust the nominal path online and effectively avoid collisions during the collaboration. • An online trajectory planning algorithm for dynamic obstacles avoidance is proposed. • A neural network CWP-net with multiple collision-free waypoints output is designed. • An improved STOMP algorithm is proposed for local trajectory optimization. • Continuous smooth trajectories are generated and satisfy joint constraints. • This work builds a bridge between advanced offline planners and online planning tasks.

Dynamic Obstacle Avoidance for USVs Using Cross-Domain Deep Reinforcement Learning and Neural Network Model Predictive Controller

This work presents a framework that allows Unmanned Surface Vehicles (USVs) to avoid dynamic obstacles through initial training on an Unmanned Ground Vehicle (UGV) and cross-domain retraining on a USV. This is achieved by integrating a Deep Reinforcement Learning (DRL) agent that generates high-level control commands and leveraging a neural network based model predictive controller (NN-MPC) to reach target waypoints and reject disturbances. A Deep Q Network (DQN) utilized in this framework is trained in a ground environment using a Turtlebot robot and retrained in a water environment using the BREAM USV in the Gazebo simulator to avoid dynamic obstacles. The network is then validated in both simulation and real-world tests. The cross-domain learning largely decreases the training time (28%) and increases the obstacle avoidance performance (70 more reward points) compared to pure water domain training. This methodology shows that it is possible to leverage the data-rich and accessible ground environments to train DRL agent in data-poor and difficult-to-access marine environments. This will allow rapid and iterative agent development without further training due to the change in environment or vehicle dynamics.

PRM-D* Method for Mobile Robot Path Planning

Various navigation tasks involving dynamic scenarios require mobile robots to meet the requirements of a high planning success rate, fast planning, dynamic obstacle avoidance, and shortest path. PRM (probabilistic roadmap method), as one of the classical path planning methods, is characterized by simple principles, probabilistic completeness, fast planning speed, and the formation of asymptotically optimal paths, but has poor performance in dynamic obstacle avoidance. In this study, we use the idea of hierarchical planning to improve the dynamic obstacle avoidance performance of PRM by introducing D* into the network construction and planning process of PRM. To demonstrate the feasibility of the proposed method, we conducted simulation experiments using the proposed PRM-D* (probabilistic roadmap method and D*) method for maps of different complexity and compared the results with those obtained by classical methods such as SPARS2 (improving sparse roadmap spanners). The experiments demonstrate that our method is non-optimal in terms of path length but second only to graph search methods; it outperforms other methods in static planning, with an average planning time of less than 1 s, and in terms of the dynamic planning speed, our method is two orders of magnitude faster than the SPARS2 method, with a single dynamic planning time of less than 0.02 s. Finally, we deployed the proposed PRM-D* algorithm on a real vehicle for experimental validation. The experimental results show that the proposed method was able to perform the navigation task in a real-world scenario.

Development of an Omnidirectional AGV by Applying ORB-SLAM for Navigation Under ROS Framework

The development of an automated guided vehicle with omni-wheels for autonomous navigation under a robot operating system framework is presented in this paper. Specifically, a laser rangefinder-constructed two-dimensional environment map is integrated with a three-dimensional point cloud map to achieve real-time robot positioning using the oriented features from accelerated segment test and rotated binary robust independent elementary feature detector-simultaneous localisation and mapping algorithm. In the path planning for autonomous navigation of the omnidirectional mobile robot, we applied the A* global path search algorithm, which uses a heuristic function to estimate the robot position difference and searches for the best direction. Moreover, we employed the time-elastic-band method for local path planning, which merges the time interval of two locations to realise time optimisation for dynamic obstacle avoidance. The experimental results verified the effectiveness of the applied algorithms for the omni-wheeled mobile robot. Further, the results showed a superior performance over the adaptive Monte Carlo localisation for robot localisation and dynamic window approach for local path planning.

A self-rotating, single-actuated UAV with extended sensor field of view for autonomous navigation.

Uncrewed aerial vehicles (UAVs) rely heavily on visual sensors to perceive obstacles and explore environments. Current UAVs are limited in both perception capability and task efficiency because of a small sensor field of view (FoV). One solution could be to leverage self-rotation in UAVs to extend the sensor FoV without consuming extra power. This natural mechanism, induced by the counter-torque of the UAV motor, has rarely been exploited by existing autonomous UAVs because of the difficulties in design and control due to highly coupled and nonlinear dynamics and the challenges in navigation brought by the high-rate self-rotation. Here, we present powered-flying ultra-underactuated LiDAR (light detection and ranging) sensing aerial robot (PULSAR), an agile and self-rotating UAV whose three-dimensional position is fully controlled by actuating only one motor to obtain the required thrust and moment. The use of a single actuator effectively reduces the energy loss in powered flights. Consequently, PULSAR consumes 26.7% less power than the benchmarked quadrotor with the same total propeller disk area and avionic payloads while retaining a good level of agility. Augmented by an onboard LiDAR sensor, PULSAR can perform autonomous navigation in unknown environments and detect both static and dynamic obstacles in panoramic views without any external instruments. We report the experiments of PULSAR in environment exploration and multidirectional dynamic obstacle avoidance with the extended FoV via self-rotation, which could lead to increased perception capability, task efficiency, and flight safety.

Development and Verification of an Autonomous and Controllable Mobile Robot Platform

In this paper, we design a mobile robot platform, which employs a fully autonomous mechanical structure and electrical control system. Two driving wheels realize flexible steering movement with four universal wheels. A variety of sensors are built on the mobile robot platform, including the Inertial Measurement Unit (IMU) used to establish the inertial navigation coordinate system and the Velodyne’s Puck lidar sensor (VLP-16) used to obtain the three-dimensional (3D) point cloud information of the environment. Then, we build a software control architecture based on the Robot Operating System (ROS), using multi-node communication to perform positioning, environment perception, dynamic obstacle avoidance, path planning and motion control. Furthermore, a method of actively exploring the environment and constructing a map is proposed, using multi-path evaluation for real-time path planning and obstacle avoidance. In the end, we conduct autonomous exploration experiments to verify the performance of the designed mobile robot platform in indoor multi-obstacle scenes.

A Novel Open-Closed-Loop Control Strategy for Quadrotor Trajectory Tracking on Real-Time Control and Acquisition Platform

This paper presents a novel closed-loop control strategy that can be applied to quadrotor trajectory tracking to improve the control accuracy and stability. Inspired by intelligent creatures, the composite control strategy combines the open-loop mode and closed-loop control to overcome some disadvantages of oscillation and overshoot under feedback control mechanisms. The open-loop mode was realized by the quadrotor maneuverability model, which is proposed to build the bridge between the flight control commands and the desired acceleration. Then, a real-time control and acquisition platform with high-precision positioning, low-delay communication, and asynchronous distribution was built to collect the UAV’s real-time flight trajectory data for training the maneuverability model and to control the quadrotors asynchronously. Based on the platform and the trained model, the composite control mode on the closed-loop of the Proportional–Integral–Derivative (PID) control and the open-loop of the Long Short-Term Memory (LSTM) maneuverability model was implemented and verified. The experimental results show that the open-loop control strategy has a better advantage in the unity of time and space over the existing techniques for improving the UAV dynamic obstacle avoidance capability.

Dynamic Obstacle Avoidance for Cable-Driven Parallel Robots With Mobile Bases via Sim-to-Real Reinforcement Learning

A Cable-Driven Parallel Robot (CDPR) with Mobile Bases (MBs) can modify its geometric architecture and is suitable for manipulation tasks in constrained environments. In manipulation tasks, a CDPR with MBs inevitably encounters obstacles, including dynamic obstacles. However, the high dimensional state space and a considerable number of constraints caused by multiple cables and MBs make the real-time dynamic obstacle avoidance of a CDPR with MBs challenging. This letter proposes a Reinforcement Learning (RL)-based dynamic obstacle avoidance method for a CDPR with MBs to deal with dynamic obstacles in real time. To explain the RL-based dynamic obstacle avoidance method, this letter focuses on a CDPR with four fixed-length cables connected to four MBs. An RL-based Obstacle Avoidance Controller (OAC) is developed and integrated into a trajectory tracking controller to address the dynamic obstacle avoidance problem of a CDPR with MBs tracking a target trajectory. To explain and evaluate the RL-based dynamic obstacle avoidance method further, an RL-based OAC is trained in a Mujoco simulator and transferred to a CDPR with four fixed-length cables connected to four MBs in the real world.

Leaning-based interfaces improve ground-based VR locomotion in reach-the-target, follow-the-path, and racing tasks.

Using standard handheld interfaces for VR locomotion may not provide a believable self-motion experience and can contribute to unwanted side effects such as motion sickness, disorientation, or increased cognitive load. This paper demonstrates how using a seated leaning-based locomotion interface --HeadJoystick-- in VR ground-based navigation affects user experience, usability, and performance. In three within-subject studies, we compared controller (touchpad/thumbstick) with a more embodied interface ("HeadJoystick") where users moved their head and/or leaned in the direction of desired locomotion. In both conditions, users sat on a regular office chair and used it to control virtual rotations. In the first study, 24 participants used HeadJoystick versus Controller in three complementary tasks including reach-the-target, follow-the-path, and racing (dynamic obstacle avoidance). In the second study, 18 participants repeatedly used HeadJoystick versus Controller (8 one-minute trials each) in a reach-the-target task. To evaluate potential benefits of different brake mechanisms, in the third study 18 participants were asked to stop within each target area for one second. All three studies consistently showed advantages of HeadJoystick over Controller: we observed improved performance in all tasks, as well as higher user ratings for enjoyment, spatial presence, immersion, vection intensity, usability, ease of learning, ease of use, and rated potential for daily and long-term use, while reducing motion sickness and task load. Overall, our results suggest that leaning-based interfaces such as HeadJoystick provide an interesting and more embodied alternative to handheld interfaces in driving, reach-the-target, and follow-the-path tasks, and potentially a wider range of scenarios.

An FTSA Trajectory Elliptical Homotopy for Unmanned Vehicles Path Planning With Multi-Objective Constraints

This article investigates a fixed-time simultaneous arrival (FTSA) problem in terms of the equilibrium of path lengths of unmanned vehicles. A novel trajectory elliptical homotopy method (TEHM) is designed to solve the FTSA problem of unmanned vehicles in a multi-objective constrained environment. Considering the constraints of obstacle avoidance and kinematics of unmanned vehicles, the trajectories elliptical homotopy is selected for path planning. The obtained trajectory homotopy guarantees obstacle avoidance and motion stability at the same time. To handle the non-cooperative and dynamic obstacle avoidance, a trajectory elliptical homotopy decomposition (TEHD) algorithm is proposed with an FTSA constraint. Based on the TEHM and TEHD, a multiple unmanned vehicle fixed-time regular-triangle formation algorithm is designed and implemented on real vehicles. Simulations and experiments validate the performance of the proposed methods and show how fixed-time arrival formation under obstacles and kinematic constraints was obtained.

Distributed dynamic obstacle avoidance design to connectivity-preserving formation control of uncertain underactuated surface vehicles under a directed network

This paper presents a Lyapunov-based recursive design approach for unifying distributed connectivity-preserving formation control and dynamic obstacle avoidance problems of uncertain multiple underactuated surface vehicles (USVs) with heterogeneous range constraints. Compared with related works, the main contribution of this study is to develop a novel control design for preserving network connectivity under dynamic obstacle avoidance in the distributed formation tracking framework. Because of the limited communication ranges, it is difficult to maintain the network interactions among USVs during dynamic obstacle avoidance. To deal with this problem, a collision-avoiding function is defined as a local error surface for the recursive control design, and a distributed complementary signal vector is introduced in the connectivity-preserving formation error surface. The complementary signals are designed to prevent network disconnections when the formation error is increased by the avoidance action and to ensure the stability of the collision-avoiding error dynamics, which eventually lead to preventing network disconnections during dynamic obstacle avoidance. An adaptive formation tracking scheme is designed via the neural networks-based command-filtered backstepping. The stability of the overall closed-loop system with preserved network connectivity and obstacle avoidance is proved based on the Lyapunov stability theorem. Finally, the theoretical approach is demonstrated through simulations.

Robust Gain-scheduling Control of Dynamic Lateral Obstacle Avoidance for Connected and Automated Vehicles

Robust Gain-scheduling Control of Dynamic Lateral Obstacle Avoidance for Connected and Automated Vehicles

Dynamic obstacle avoidance and trajectory planning of five-axis redundant industrial manipulator

Aiming at the problem that five axis redundant industrial manipulator dynamic obstacle avoidance and trajectory planning algorithm does not consider the minimum difference of each joint of the manipulator, which leads to low success rate of obstacle avoidance planning, slow convergence speed of path cost and long time of obstacle avoidance planning, a simulation study on dynamic obstacle avoidance trajectory planning of five axis redundant industrial manipulator is proposed. According to the D-H rule, the coordinate system of each link joint of the five axis redundant industrial manipulator is established, and the forward and inverse kinematics of the five axis redundant industrial manipulator is analyzed. AABB's hierarchical bounding box tree algorithm is used to detect the collision of five axis redundant industrial manipulator. This paper uses harmony search algorithm to plan the obstacle avoidance path of five axis redundant industrial manipulator, determines the objective function and constraints of the optimization problem, sets algorithm parameters, initializes harmony memory, creates new harmony, updates harmony memory, checks and searches the target state, achieves the maximum number of iterations, and realizes the dynamic obstacle avoidance and trajectory planning of five axis redundant industrial manipulator. The experimental results show that the path cost of the proposed algorithm converges faster, and can effectively improve the success rate of obstacle avoidance planning and shorten the time of obstacle avoidance planning.

Research on Autonomous Navigation of Showroom Service Robot Based on ROS and Laser SLAM

As an important tool for guiding and introducing visitors to the exhibition hall, the robot needs to confront the complex environment that may exist and the constantly moving visitors. First of all, it is needed to have the ability to enable accurate positioning and navigation as well as dynamic obstacle avoidance. In recent years, simultaneous localization and mapping (SLAM) technology has been widely used to address localization and navigation issues in unknown indoor environments. Based on the robot operating system (ROS), this study collects environmental information by 2D LiDAR, and then combines the service robot's Inertial measurement unit (IMU), odometer, and other sensors to map the exhibition hall environment by the Cartographer SLAM algorithm. Path planning and dynamic obstacle avoidance functions are also implemented, in path planning, the algorithm is selected for global path planning and the Dynamic Window Approach (DWA) algorithm for local path planning. The environment map construction, path planning and dynamic obstacle avoidance functions are verified by simulation and in real environment. The average position deviation and standard deviation (SD) of the robot from the target point were less than 7 cm and less than 3 cm, respectively, when the robot was set to move at 0.5 m/s for the positioning test. The average heading deviation was less than 9° and SD was less than 3°. The robot's positioning and navigation can well meet the working requirements of the pavilion service robot.