面向机器人笛卡尔空间的轨迹运动自适应抑振控制

Adaptive vibration suppression control for trajectory motion in Cartesian space of industrial robot

工业机器人的柔性传动部件导致机器人末端产生振动,严重降低笛卡尔运动精度,为此提出自适应输入整形抑振控制策略。以"双惯量模型+PD控制"建立关节伺服系统模型,分析振动特性。由于运动中的振动信号存在较大噪声,无法准确辨识振动频率和阻尼比,故采用基于递推最小二乘法(Recursive Least Squares,RLS)的任意时延输入整形器自适应设计方法。通过对笛卡尔轨迹规划的归一化插补函数输入整形,提出笛卡尔轨迹输入整形方法,实际表现为重新规划笛卡尔速度大小而不改变运动方向。输入整形抑振实验中直线轨迹振动信号能量最大降低77.08%,圆弧轨迹振动信号能量最大降低31.07%,残余振动信号能量降低80%以上,验证了所述抑振控制策略能有效抑制机器人笛卡尔运动启停时的振动。

The flexible transmission parts cause vibration at the end of industrial robots,which severely reduces accuracy of Cartesian motion so the adaptive input shaping control strategy is proposed.Servo system model of flexible joint is founded by‘two-inertia model+PD control’and vibration characteristic is discussed.It is hard to accurately identify frequency and damping ratio,as a consequence of noise problem of vibration signal collected during Cartesian motion.Therefore,the adaptive design method of arbitrary delay input shaper based on the recursive least squares method(RLS)is adopted.By performing input shaping on the normalized interpolation function of Cartesian trajectory planning,a Cartesian trajectory input shaping method is proposed.The actual performance is replanning the Cartesian velocity without changing the direction.In the vibration suppression experiment,vibration signal energy of linear and arc motion is reduced by up to 77.08%and 31.07%respectively while residual vibration signal energy is reduced by more than 80%,which is verified that the control strategy can effectively suppress the vibration in the Cartesian motion of industrial robots.