An Autonomous Obstacle Avoidance Path Planning Method Involving PSO for Dual-Arm Surgical Robot

Qiao Chen,Yiwei Liu,Zhuo Chen,Yangjunjian Zhou
DOI: https://doi.org/10.1109/icmra56206.2022.10145637
2022-01-01
Abstract:Currently, motion planning and reasonable obstacle avoidance of surgical robots are essential research directions. Most surgical robots adopt a simple master-slave control strategy and cannot avoid obstacles autonomously. To this end, we propose an autonomous obstacle avoidance method for the dual-arm surgical robot. Firstly, the obstacle avoidance path planning is based on the artificial potential field method. We propose the splicing algorithm to make the manipulator jump out of the local minimum and improve the planning efficiency. And using the adaptive step to stable motion. Secondly, for collision detection, we use the Gilbert-Johnson-Keerthi (GJK) algorithm to calculate the distance between the obstacle and the equivalent convex body of the manipulator. A warning appears when there is a collision. Because the artificial potential field method lacks time planning, we use the time-optimized 3-5-3 polynomial interpolation based on the Particle Swarm optimization (PSO) to the obtained path. And make manipulators perform the surgery without collision for a suitable period. Finally, we carry out the system simulation, physical experiments, and dual-arm obstacle avoidance experiments based on the above autonomous obstacle avoidance algorithm. Experiments show that this study can make the dual-arm surgical robot move quickly and smoothly on the premise of avoiding obstacles, which provides a reference for the automation of surgical robots.
What problem does this paper attempt to address?