• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2020, Vol. 56 ›› Issue (17): 59-70.doi: 10.3901/JME.2020.17.059

• 机器人及机构学 • 上一篇    下一篇

扫码分享

基于伪距离的冗余机器人避障算法

万俊1, 姚佳烽2, 余亮3, 张良安4, 吴洪涛2   

  1. 1. 江苏理工学院汽车与交通工程学院 常州 213001;
    2. 南京航空航天大学机电学院 南京 210016;
    3. 淮南师范学院机械与电气工程学院 淮南 232038;
    4. 安徽工业大学机械工程学院 马鞍山 243032
  • 收稿日期:2019-10-16 修回日期:2020-06-09 出版日期:2020-09-05 发布日期:2020-10-19
  • 通讯作者: 万俊(通信作者),男,1989年出生,讲师,博士。主要研究方向为协作机器人,机器人柔顺控制技术。E-mail:junwanrob@gmail.com
  • 基金资助:
    国家自然科学基金资助项目(51706098)。

Obstacle Avoidance Algorithm for Redundant Manipulators Based on Pseudo-distance Method

WAN Jun1, YAO Jiafeng2, YU Liang3, ZHANG Liangan4, WU Hongtao2   

  1. 1. School of Automobile and Traffic Engineering, Jiangsu University of Technology, Changzhou 213001;
    2. College of Mechanical and Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016;
    3. College of Mechanical and Electrical Engineering, Huainan Normal University, Huainan 232038;
    4. Department of Mechanical Engineering, Anhui University of Technology, Maanshan 243032
  • Received:2019-10-16 Revised:2020-06-09 Online:2020-09-05 Published:2020-10-19

摘要: 针对冗余机器人避障问题,提出一种基于伪距离且无需主从任务转换的冗余机器人避障算法。采用具有解析式的伪距离算法计算机器人与障碍物之间的接近程度,通过分析伪距离与欧式距离换算关系求取机器人与障碍物之间最小欧式距离,从而提高冗余机器人与障碍物之间的最小欧式距离求解效率。基于最小伪距离判别指标,在机器人末端轨迹上赋予末端沿最小距离方向避障速度实现末端避障。在零空间避障问题求解中提出一种自适应阻尼控制方法,产生虚拟排斥力使机器人连杆远离障碍物,并自适应调整阻尼增益参数响应避障速度。通过七自由度冗余机器人的仿真和试验验证了该算法的高效性和有效性。仿真结果表明,该算法提出的末端避障方法避免了传统的主从任务转换避障算法在任务切换时所引起的末端振荡且缩短了避障时间,同时该算法提出的零空间避障方法能够在低误差跟踪末端期望轨迹下避免机器人连杆与障碍物相碰撞。试验结果表明,该算法能够完成多障碍物避障,验证了该算法的末端避障方法和零空间避障方法的正确性。

关键词: 冗余机器人, 伪距离, 末端避障, 零空间避障

Abstract: Focusing on the obstacle avoidance problem of redundant manipulators, an algorithm based on pseudo-distance technique for avoiding obstacles in redundant robot control is presented without master-slaver task conversion. The proximity between the redundant manipulator and the obstacle is calculated by using the pseudo-distance algorithm with analytic form. Then the minimum Euclidean distance is obtained by analyzing the conversion relation of the pseudo-distance and Euclidean distance, leading to resolving the minimum Euclidean distance efficiently between the manipulator and the obstacle. On the one part, based on the identification index of minimum pseudo-distance, obstacle avoidance at end-effector of manipulator is accomplished by endowing a repelling velocity of end-effector along the direction of minimum Euclidean distance. On the other hand, an adaptive damping control method based on the identification index is proposed for obstacle avoidance in null space of a redundant robot, generating repelling virtual force to push robot links away from obstacles and adjusting damping factors adaptively responding to repelling force. Finally, the efficiency and effectiveness of the proposed algorithm is verified by simulations and experiments of a 7 Degree-of-Freedom (DOF) redundant manipulator. The simulation results show the oscillation of end-effector caused by the master-slave task transformation of the traditional obstacle-avoiding method is avoided and obstacle avoidance running time is reduced dramatically in the end-effector obstacle avoidance, and in the null space obstacle avoidance, the avoidance of collisions between the manipulator and the obstacle is realized successfully under the low error tracking the desired trajectory. The final experiment results show that the multiple obstacles avoidance is also implement splendidly based on the proposed obstacle avoidance strategy and the correctness of the designed end-effector obstacle avoidance method and the null space obstacle avoidance method is further demonstrated.

Key words: redundant manipulator, pseudo-distance, end-effector obstacle avoidance, null space obstacle avoidance

中图分类号: