基于旋转势场法的双足机器人避障路径规划方法  

Path Planning Method Based on Rotary Potential Field for a Biped Robot

作　　者：宋晖[1,2] 殷国富[1] 李光日[3] 

机构地区：[1]四川大学制造科学与工程学院,四川成都610065 [2]西南科技大学计算机学院,四川绵阳621000 [3]中国空间技术研究院,北京100094

出　　处：《四川大学学报（工程科学版）》2015年第5期167-171,共5页Journal of Sichuan University (Engineering Science Edition)

基　　金：国家科技重大专项资助项目(2013ZX04005-012);四川省科技支撑计划资助项目(2012GC0008)

摘　　要：针对双足机器人在实际环境中运动时难以实时规避障碍物的问题,提出一种基于旋转势场法的避障路径规划方法。该方法将足迹规划思想引入基于势场法的局部实时路径规划中,通过设计新的障碍物旋转势场来解决传统势场法的局部极小值问题,进而利用旋转势场和足迹规划间的映射关系实现双足机器人的实时避障运动。将该算法应用到1台双足机器人上进行实验验证,现场设定障碍物使路径更接近实际作业环境,机器人顺利完成了U型场地的避障,验证了该方法的有效性。Since the biped robots are difficult to avoid obstacles in real time due to the balance constraints when the biped robots walk in actual environments, a path planning and obstacle avoidance method was proposed based on the rotating potential field. First, the idea of footprint planning was introduced into the local real-time path planning based on the potential field. Then, the local minima problem in the traditional potential field was solved by designing a new obstacle rotation potential field and the real-time obstacle avoidance was achieved using the mapping relationship. Finally, the effectiveness of the proposed method was confirmed by obstacle avoidance experiments on a small size biped robot in the environment with U-shaped obstacles.

关 键 词：双足机器人 旋转势场 避障 路径规划 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]