A Knee-Guided Evolutionary Algorithm Based Navigation Approach for Mobile Robots in Intelligent Manufacturing Scenarios

Aiming at the problem of multi-objective path planning (MOPP) and dynamic obstacle avoidance of manipulator in dynamic environment, this paper proposes a solution based on hierarchical reinforcement learning (HRL) framework. Traditional path planning methods have some problems in dynamic scenes, such as poor real-time performance, difficult to balance multi-objective conflicts and insufficient adaptability to environmental changes. Therefore, this paper designs a two-tier architecture including global path planning layer and local obstacle avoidance layer, which is trained by Proximal policy optimization (PPO) and Soft Actor-Critic (SAC) algorithms respectively, and achieves collaborative optimization through an efficient information exchange mechanism between layers. At the same time, a multi-objective reward function based on dynamic weight adjustment strategy is introduced. Fuzzy logic is used to adaptively balance the relationship among path length, obstacle avoidance safety and energy consumption according to environmental complexity. Combined with Long Short-Term Memory (LSTM), the trajectory of obstacles is predicted, and the potential field method is further introduced to modify the obstacle avoidance reward, which improves the real-time response ability and robustness of the algorithm in dynamic environment. The experimental results show that the HRL-SAC-PPO method proposed in this paper shows superior performance in both static and dynamic scenarios. In the static scene, the success rate of this method reaches 100%, the average path length is shortened to 2.13m, no collision occurs, and the energy consumption is reduced to 1.12 kJ, which shows a good multi-objective optimization effect. In the dynamic scene, the trajectory error of obstacles predicted by LSTM is only 4.2%, and the safe distance between the robot arm and obstacles is improved by 35%, which significantly enhances the reliability of obstacle avoidance. In addition, the average decision delay of this method is only 11.3ms, and the peak delay is 23ms, which is much lower than that of the contrast algorithm, showing stronger real-time response ability. The ablation experiment further verified the key role of LSTM trajectory prediction, dynamic weight adjustment and layered structure on the overall performance. In the welding task verification of the real UR5 manipulator, the success rate of the system in dynamic environment is 95.3%, the average path length is 3.41m, and the maximum joint acceleration is only 0.87 radian/s², which is far below the safety threshold, indicating that the algorithm has good stability and obstacle avoidance ability in practical application. The comprehensive performance comparison shows that this method performs well in different industrial scenarios, and has stronger environmental adaptability and comprehensive path planning ability.

The traditional Rapidly-exploring Random Tree Star (RRT*) suffers from the low path generation efficiency, numerous invalid exploration points, and unsuitability for navigation in unknown dynamic environments. In this article, we propose a dynamic path planning scheme by combining the improved RRT* and the improved dynamic window approach (DWA). For pregenerating an initial path, we use the artificial potential field (APF) method to expand new nodes. The adaptive dynamic step-size is introduced for accelerating the optimal path searching. Moreover, the improved ant colony algorithm is used to perform multiobjective optimization on the generated initial path. When unknown obstacles appear in the path, the improved DWA is developed for obstacle avoidance. Finally, the proposed method is validated by simulation and experiment in both of the static and dynamic environments. In particular, the simulation results show that, compared with some existing methods, our algorithm can generate a higher-quality initial path in the static environment and avoid unknown dynamic obstacles effectively in the dynamic environment. Furthermore, we implement our algorithm in a mobile robot to verify the correctness and effectiveness in the practical scenario.

This paper focuses on obstacle avoidance in the source location problem, in which robots capture the signal strength and find the signal source in an unknown environment. This work proposes a particle swarm optimizer (PSO) with a hybrid obstacle avoidance strategy to solve the problem. The signal strength is considered as the fitness function for PSO to guide robots. During moving, artificial potential fields are adopted to make robots avoid obstacles and each other. A deadlock escaping strategy is put forward to deal with the constraints of concave obstacles. The weighted average velocity of a robot is employed to check whether it is stuck by an obstacle. If so, a tabu area is set to push robots out of the area and prevent them from searching the same place again. These tabu areas offer robots key information about obstacles in an unknown environment and improve robots' ability of obstacle avoidance. The proposed algorithm is adaptive in unknown environments, meaning that no prior knowledge is needed. Simulation tests verify the effectiveness of the developed algorithm, showing satisfactory performance when dealing with concave obstacles.

This paper presents an obstacle avoidance solution for an existing power line inspection robot (PLIR) prototype. The PLIR's workspace environment is represented by a discretised 3D map, which is used for obstacle avoidance manoeuvre planning for both known and unknown obstacles. Obstacles are represented in this workspace using a union of axis aligned bounding boxes. The Lazy Theta* algorithm is then used to find the shortest 3D graph path around the obstacle. The resulting path nodes are used to generate a B-spline trajectory. The trajectory is simulated to check for collisions and, if collision free, is used to command the actual robot. Experimental results show that this obstacle avoidance strategy is successful, with the prototype robot being able to manoeuvre around an obstacle in laboratory tests.

The obstacle avoidance controller is a key autonomous component which involves the control of tractor system dynamics, such as the yaw lateral dynamics, the longitudinal dynamics, and nonlinear constraints including the speed and steering angles limits during the path-tracking process. To achieve the obstacle avoidance ability of control accuracy, an independent path re-planning controller is proposed based on ROS (Robot Operating System) nonlinear model prediction in this paper. In the design process, the obstacle avoidance function and an objective function are introduced. Based on these functions, the obstacle avoidance maneuvering performance is transformed into a nonlinear quadratic optimization problem with vehicle dynamic constraints. Moreover, the tractor dynamics maneuvering performance can be effectively adjusted through the proposed objective function. To validate the proposed algorithm, a ROS based tractor dynamics model and the SLAM (Simultaneous Localization and Mapping) are established for numerical simulations under different speed. The maximum obstacle avoidance deviation in the simulation is 0.242 m at 10 m/s, and 0.416 m at 30 m/s. The front-wheel rotation angle and lateral velocity are within the constraint range during the whole tracking process. The numerical results show that the designed controller can achieve the tractor obstacle avoidance ability with good accuracy under different conditions. Keywords: ROS, obstacle avoidance, nonlinear model prediction, agricultural tractor DOI: 10.25165/j.ijabe.20191206.4907 Citation: Liu Z D, Lv Z Q, Zheng W X, Zhang W Z, Cheng X X. Design of obstacle avoidance controller for agricultural tractor based on ROS. Int J Agric & Biol Eng, 2019; 12(6): 58–65.

Obstacle avoidance ability is the significant embodiment of the ground mobile robot, and the basic guarantee of the ground mobile robot to perform various tasks. Obstacle avoidance technologies are divided into two kinds, one is based on the global map and another is based on sensors respectively. This paper mainly aims at the local obstacle avoidance method based on sensors. The study of obstacle detection and obstacle avoidance are two inseparable parts in the research of obstacle avoidance ability. This paper proposes an efficient obstacle detection and obstacle avoidance algorithm based on 2-D lidar. A method is proposed to get the information of obstacles by filtering and clustering the laser-point cloud data. Also, this method generates the forward angle and velocity of robot based on the principle of minimum cost function. The obstacle detection and obstacle avoidance algorithm has advantages of a simple mathematical model and good real-time performance. The effectiveness of the proposed algorithm is verified on MATLAB simulation platform.

With the booming development of logistics, manufacturing and warehousing fields, the autonomous navigation and intelligent obstacle avoidance technology of automated guided vehicles (AGVs) has become the focus of scientific research. In this paper, an enhanced deep reinforcement learning (DRL) framework is proposed, aiming to empower AGVs with the ability of autonomous navigation and obstacle avoidance in the unknown and variable complex environment. To address the problems of time‐consuming training and limited generalisation ability of traditional DRL, we refine the twin delayed deep deterministic policy gradient algorithm by integrating adaptive noise attenuation and dynamic delayed updating, optimising both training efficiency and model robustness. In order to further strengthen the AGV's ability to perceive and respond to changes of a dynamic environment, we introduce a distance‐based obstacle penalty term in the designed composite reward function, which ensures that the AGV is capable of predicting and avoiding obstacles effectively in dynamic scenarios. Experiments indicate that the AGV model trained by this algorithm presents excellent autonomous navigation capability in both static and dynamic environments, with a high task completion rate, stable and reliable operation, which fully proves the high efficiency and robustness of this method and its practical value.

In this paper, a novel obstacle avoidance algorithm for the multi-UAV flocking was proposed by combining artificial potential field and Dubins path planning method. For the multi-UAV flocking, a virtual-leader was considered, and its flight path was designed by Dubins path planning method, which can take into account the overload constraint of the UAV. In this way, the flight path of the whole multi-UAV flocking was roughly planned, so as to save obstacle avoidance time, and the flight path of the follower-UAVs was designed according to the artificial potential field method, which has a better performance in real-time obstacle avoidance, and the algorithm is relatively simple. The simulation results show that the obstacle avoidance algorithm designed by combining the two methods has a better obstacle avoidance ability.

As human–robot collaboration increases tremendously in real-world applications, a fully autonomous and reliable mobile robot for the collaboration has been a central research topic and investigated extensively in a large number of studies. One of the most pressing issues in such topic is the collision-free navigation that has a moving goal and unknown obstacles under the unstructured environment. In this article, a cognitive robotic system (CRS) is proposed for the robot to navigate itself to the moving target person without obstacle collision. This CRS consists of a cognitive agent, which is created based on the Soar cognitive architecture to reason its current situation and make action decision for the robot to avoid obstacles and reach the target position, and a speed planning module, which is based on dynamic window approach (DWA) to generate appropriate linear and angular velocities for driving the robot’s motors. For the implementation of the proposed system, we use a differential drive wheel robot equipped with two ultrawideband (UWB) sensors and a color depth camera as the experimental platform. Finally, to evaluate the performance of our system in actual operating conditions, we conduct experiments with a scenario that includes main tasks: avoiding consecutive unknown obstacles and turning at corner while the robot follows continuously human user along the corridor.

With the robot gradually stepping into our daily life, more and more attention has been paid to the mobile ability of robot, and obstacle avoidance is a key problem. In this paper, the indoor environment is taken as the application scene, and the visual detection and local dynamic obstacle avoidance are studied respectively. Aiming at the shortcomings of the single external sensor lidar’s incomplete perception of obstacle information, this paper proposes an obstacle avoidance method based on machine vision for mobile obstacle detection under the ROS operating system. The method is improved based on YOLO-v4 in terms of vision, which can meet the real-time requirements of mobile terminal. Combined with dynamic window approach (DWA), the local obstacle avoidance algorithm is improved. In the process of local obstacle avoidance, visual detection information is integrated to increase the ability of local dynamic obstacle avoidance and improve the performance of robot local obstacle avoidance. Finally, the feasibility and validity of the algorithm are verified in the actual environment.

In this paper，a novel method for obtaining high-quality paths for self-driving cars in underground parking lots is proposed. Self-driving cars require fast and accurate planning of collisionless path. When the self-driving car arrives at the parking lot, the car downloads the layout from the intelligent system of the parking lot and is assigned a parking space, then the location of the designated parking space and the car are provided by the intelligent system. A global path is planned by the global algorithm according to the location of the parking space and the car as well as the layout. If dynamic or unknown obstacles are detected in the process of moving along the global path, the parameters of obstacles can be estimated by the obstacle-detection algorithm. According to obtained parameters, the local obstacle avoidance path can be planned by the behavioral dynamics method. After completing obstacle avoidance, then the car will return to the global path and continue to move toward the target parking space. Finally, the proposed method is simulated by MATLAB, and the results show that the car can safely park in the target parking space. This method simultaneously satisfies the smooth and the real-time requirements of path planning.

Today, real-time collision detection is a basic demand for many applications. While collision tests between known (modeled) objects have been around for quite a while, collision detection of known objects with dynamic, unknown (sensor-detected) objects remains a challenging field of research, especially when it comes to real-time requirements. The collision test described in this paper is based on several stationary, calibrated video cameras, each supervising the entire 3-dimensional space shared by unknown and known objects (e.g. humans and robots). Based on their images, potential collisions of the known objects in any of their (future) configurations with a priori unknown dynamic obstacles are detected. Occlusions caused by known objects (such as the robot or machinery set-up within the workspace) are detected and addressed in a safe manner by exploiting the geometrical information of the known objects and the epipolar line geometry of the calibrated cameras in a decision fusion process. The algorithm can be parameterized to adapt to different application demands. Experimental validation shows that real-time behaviour is possible in the presence of highly dynamic unknown obstacles as they occur when humans and robots share the same workspace for the accomplishment of a shared task. In effect, the vision-based collision test can safely be used for human-robot cooperation, intrusion detection, velocity damping, or obstacle avoidance.

In this paper, multiple autonomous underwater vehicle (multi-AUV) formation control with obstacle avoidance ability in 3D complex underwater environments based on an event-triggered model predictive control (EMPC) is proposed. Firstly, multi-AUV motion model systems are developed. The navigation reference trajectory of the follower AUVs can be obtained using a multi-AUV relative motion model. Secondly, in order to overcome the speed jump and obstacle avoidance problem in multi-AUV systems, compatibility constraints are presented in MPC that limit the uncertainty deviation of each AUV. The event-triggered mechanism (ET) is designed to decrease the computational load, which is based on the error between the optimal predicted and current state of the AUV. Finally, the effectiveness and superiority of the proposed algorithm are confirmed via simulation and compared with those of other algorithms.

Objectives In order to solve the problem that unmanned boats cannot carry out reasonable and effective autonomous obstacle avoidance under dense obstacle sea conditions. Methods An improved dual-window dynamic window approach（DWA）obstacle avoidance algorithm based on fuzzy control strategy is proposed, that is, a sensing window based on the onboard sensor is designed on the basis of the conventional speed window, and the dual-window model composed of the speed window and the sensing window is further optimized The speed space is constrained, and the weight of the evaluation function is dynamically adjusted based on the fuzzy control strategy based on the obstacle distribution state and the distance between the UAV and the obstacle. Results The simulation experiment results show that, compared with the original DWA algorithm, the path planned by the improved algorithm under unknown dense obstacle sea conditions is more reasonable and smoother, which not only solves the problem of unmanned boats detouring outside the dense obstacle group, but also takes into account The safety of obstacle avoidance navigation is improved, the number of iterations and running time can be shortened by more than 20%. Conclusions The research results have certain reference value for the research on autonomous obstacle avoidance technology of unmanned boats.

PurposeWheeled mobile robots (WMR) are the most widely used robots. Avoiding obstacles in unstructured environments, especially dynamic obstacles such as pedestrians, is a serious challenge for WMR. This paper aims to present a hybrid obstacle avoidance method that combines an informed-rapidly exploring random tree* algorithm with a three-dimensional (3D)-object detection approach and model prediction controller (MPC) to conduct obstacle perception, collision-free path planning and obstacle avoidance for WMR in unstructured environments.Design/methodology/approachGiven a reference orientation and speed, the hybrid method uses parametric ellipses to represent obstacle expansion boundaries based on the 3D target detection results, and a collision-free reference path is planned. Then, the authors build on a model predictive control for tracking the collision-free reference path by incorporating the distance between the robot and obstacles. The proposed framework is a mapless method for WMR.FindingsThe authors present experimental results with a mobile robot for obstacle avoidance in indoor environments crowded with obstacles, such as chairs and pedestrians. The results show that the proposed hybrid obstacle avoidance method can satisfy the application requirements of mobile robots in unstructured environments.Originality/valueIn this study, the parameter ellipse is used to represent the area occupied by the obstacle, which takes the velocity as the parameter. Therefore, the motion direction and position of dynamic obstacles can be considered in the planning stage, which enhances the success rate of obstacle avoidance. In addition, the distance between the obstacle and robot is increased in the MPC optimization function to ensure a safe distance between the robot and the obstacle.