基于力感知与GPF-RRT*算法的机器人主从控制研究

Research on master-slave control of robots based on force perception and GPF-RRT * algorithm

摘要:

研究对机器人的主从控制方法进行了分析，发现目前主从控制方法均存在效率低下，可控度低等问题。因此，研究采用了优化随机树算法结合高斯函数势场路径规划方法并辅以力感知的关键技术对主从控制约束进行分析。实验结果显示，采样优化后的路径平均规划时间远低于采样优化前的时间，优化后的时间曲线在第40次迭代后呈现下降趋势。采样优化后的路径平均规划长度在第50次迭代后呈现下降趋势，50次迭代前呈现稳步上升趋势。这说明了研究所设计控制系统能够准确、高效地对机器人进行控制，实现了最低程度的反馈误差。最终经仿真应用满足了机器人行业亟需的主从控制需求，进一步提升了机器人的可操作性。

Abstract:

This article studies the master-slave control methods of robots and finds that current master-slave control methods have problems such as low efficiency and low controllability. Therefore, the study adopted an optimized random tree algorithm combined with Gaussian function potential field path planning method and supplemented with key techniques of force sensing to analyze the constraints of master-slave control. The experimental results show that the average planning time of the path after sampling optimization is much lower than that before sampling optimization, and the optimized time curve shows a downward trend after the 40th iteration. The average planning length of the path after sampling optimization showed a downward trend after the 50th iteration, and a steady upward trend before the 50th iteration. This indicates that the control system designed by the research institute can accurately and efficiently control the robot, achieving the lowest level of feedback error. The final simulation application met the urgent need for master-slave control in the robotics industry, further enhancing the operability of robots.