引用本文:张雪,刘成菊,陈启军.基于Fuzzy–CPG的双足机器人适应性行走控制[J].控制理论与应用,2020,37(12):2525~2534.[点击复制]
ZHANG Xue,LIU Cheng-ju,CHEN Qi-jun.Adaptive walking control of biped robot based on Fuzzy-CPG[J].Control Theory and Technology,2020,37(12):2525~2534.[点击复制]
基于Fuzzy–CPG的双足机器人适应性行走控制
Adaptive walking control of biped robot based on Fuzzy-CPG
摘要点击 2045  全文点击 626  投稿时间:2020-02-13  修订日期:2020-06-29
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2020.00076
  2020,37(12):2525-2534
中文关键词  双足机器人  模糊控制  中枢模式发生器(CPG)  适应性行走
英文关键词  biped robot  fuzzy control  CPG(central pattern generator)  adaptive walking
基金项目  国家自然科学基金;上海市科委基础研究项目
作者单位E-mail
张雪 同济大学 zhangxue@tongji.edu.cn 
刘成菊* 同济大学 liuchengju@tongji.edu.cn 
陈启军 同济大学  
中文摘要
      为提高双足机器人的环境适应性, 本文提出了一种基于模糊控制与中枢模式发生器(CPG)的混合控制策 略, 称之为Fuzzy–CPG算法. 高层控制中枢串联模糊控制系统, 将环境反馈信息映射为行走步态信息和CPG幅值参 数. 低层控制中枢CPG根据高层输出命令产生节律性信号, 作为机器人的关节控制信号. 通过机器人运动, 获取环境 信息并反馈给高层控制中枢, 产生下一步的运动命令. 在坡度和凹凸程度可变的仿真环境中进行混合控制策略的 实验验证, 结果表明, 本文提出的Fuzzy–CPG控制方法可以使机器人根据环境的变化产生适应的行走步态, 提高了 双足机器人的环境适应性行走能力.
英文摘要
      To improve the walking adaptive ability of biped robot, a hybrid control strategy (Fuzzy-CPG) based on fuzzy control and CPG(central pattern generator) is proposed. The environmental feedback information is mapped into expected walking gait information of the robot and amplitude parameters of the CPG model through the high-level control center, series fuzzy control system. According to the output command of the high-level control center, rhythmic output signals controling the joint motion of the robot are generated by the low-level control center CPG. The robot detects the environment feedback information and feeds it back to the high-level control center, where further motion commands are generated to start a new cycle. In simulation environments with variable slope angles and unevenness, the hybrid control strategy is verified. The results show that the proposed hybrid control strategy (Fuzzy-CPG) can generate corresponding walking gaits according to the change of environment and improve the adaptive walking ability of biped robot.