Dynamic obstacle avoidance path planning for collaborative robot based on improved APF-Bi-RRT algorithm

In order to enable a collaborative robot to perform path planning and operate more smoothly in environments with dynamic obstacles, a dynamic obstacle avoidance algorithm, named improved APF-Bi-RRT, is proposed by combining an improved Artificial Potential Field (APF) and an improved Bidirectional Rapidly-exploring Random Tree (Bi-RRT). First, a cost function is constructed with the goal of minimizing the angle change of each joint of the robot to select path nodes planned by the Bi-RRT algorithm, reducing redundant nodes and improving path quality. Second, considering the influence of dynamic obstacles on the various links of the robot, the gravitational field function of the artificial potential field method is constructed in the joint space, and the repulsive field function is reconstructed. Finally, the two algorithms are combined, and simulations are conducted in environments with static and dynamic obstacles. The results show that the improved APF-Bi-RRT algorithm enables the collaborative robot to achieve dynamic obstacle avoidance with smooth paths and smooth operation of all joints, which verifies the effectiveness of the algorithm.