摘要
提出一种视觉引导的面向位控机器人的力/位混合控制策略。结合视觉传感器的测量特点,对未知环境边缘进行估计获得位控和力控方向,根据位控和力控方向在线对机器人终端的运动轨迹进行规划,并采用阻抗控制规律使机器人获得较好的柔顺性。在该算法中应用一个参考比例因子调节参考轨迹,根据反馈的接触力信息通过模糊推理确定参考比例因子的大小,从而使生成的参考轨迹适应未知环境刚度的变化。在小视场条件下(26.4mm×26.4mm)跟踪速度达到15mm/s,并且具有较高的力控制精度。
A hybrid force/position control strategy for position-controlled robot was presented based on vision-guided.The edge of unknown environment was estimated to gain the vectors of position control and force control.The estimated vectors were used to generate the virtual reference trajectory of the robot real time,and impedance control rule was used to make the robot gain good compliance.A reference scaling factor was developed to tune the reference trajectory.The value of this factor was determined by fuzzy reasoning in accordance to the feedback information of forces in contact in order to adapt the reference trajectory generated to the changeable unknown stiffness in environment.The Experimental results show the tracking speed reaches 15 mm/s with a small field of view（26.4 mm×26.4 mm）,and its force tracking capability is verified.
出处
《机床与液压》
北大核心
2010年第11期10-11,6,共3页
Machine Tool & Hydraulics
基金
国家自然科学基金资助项目(30670529)
兰州理工大学电气与控制学科团队基金资助
关键词
视觉引导
力/位混合控制
阻抗控制
未知环境
Vision-guided
Hybrid force/position control
Impedance control
Unknown environment
