A Repetitive Motion Obstacle Avoidance Planning Scheme for Omnidirectional Mobile Manipulators with Noise Suppression Capability

Abstract

Obstacle avoidance is essential for an omnidirectional mobile robotic arm (OMRA) to accomplish a given task in a complex environment. To make OMRA accomplish repetitive motion, obstacle avoidance, and trajectory tracking tasks in complex dynamic environments with the presence of obstacles and external disturbances, a new noise-suppressed repetitive motion obstacle avoidance (NSRMOA) scheme is proposed. As established previously many OMRA motion planning assumes that the simulation is noise free. However, noise interference may prevent OMRA from accomplishing repetitive motion, obstacle avoidance, and trajectory tracking tasks, which may lead to task failure. The proposed the NSRMOA scheme with intrinsic noise suppression can solve external noise interference, so that the OMRA can accomplish the repetitive motion, obstacle avoidance and trajectory tracking tasks in the presence of external noise interference. Finally, the optimal solution of the NSRMOA scheme can be effectively obtained by using recurrent neural network (RNN). The effectiveness and accuracy of the scheme are verified by examples and comparisons.