Optimization method of obstacle avoidance path for dual-arm cooperative robot based on machine vision

This study focuses on optimizing the obstacle avoidance path for a dual-arm cooperative robot using machine vision techniques, aiming to ensure that the robot can navigate around obstacles smoothly and swiftly. Machine vision technology is applied to enhance the obstacle avoidance path planning of the dual-arm cooperative robot. Specifically, a binocular vision camera is mounted on the robot’s head to capture images of the driving path. Zhang Zhengyou’s calibration method is employed to calibrate the binocular vision camera, determining the camera model parameters. The Bouguet algorithm is then used to rectify the obstacle avoidance path images captured by the binocular vision camera. Machine vision methods are utilized to identify both static and dynamic obstacles along the driving path of the dual-arm cooperative robot. The motion model of the robot is analyzed, and a fitness function for the obstacle avoidance path is constructed. An improved bat algorithm is employed to optimize the obstacle avoidance path of the dual-arm cooperative robot, ultimately yielding the optimal path. Experimental results demonstrate that the re-projection error of this method is less than 0.09 pixels. By calibrating the binocular vision camera using Zhang Zhengyou’s method, optimal camera parameters are obtained, enhancing the calibration accuracy of the binocular vision camera. Furthermore, this approach effectively identifies both static and dynamic obstacles, successfully avoiding newly added obstacles while maintaining the shortest possible path length, thereby achieving optimal obstacle avoidance path planning for the dual-arm cooperative robot.