基于卡尔曼滤波的机器人自适应控制方法研究

Kalman filter-based adaptive control method for robots

针对机械臂在工作过程中与环境交互的力柔顺控制问题,提出了一种基于卡尔曼滤波的自适应阻抗控制算法,完成了在环境位置与环境刚度未知的情况下对恒力的稳定控制。首先,构建了机械臂与环境接触的阻抗控制模型,建立了力与位置之间的关系;然后,建立了自适应阻抗控制模型,根据机械臂力跟踪误差与位置误差,对机械臂接触的环境位置和环境刚度进行了估计,对机械臂控制位置进行了更新,再将阻抗控制算法离散化,并利用卡尔曼滤波对阻抗控制系统状态进行了预估和校正;最后,仿真验证了自适应卡尔曼滤波阻抗模型的稳定性与有效性,并运用机器人进行了跟踪实验,对不同环境参数下的恒力跟踪效果进行了验证。研究结果表明:基于卡尔曼滤波的自适应阻抗控制算法对力跟踪误差可以达到5%以内,在斜面环境下的收敛速度可提高61.5%;与传统阻抗控制相比,该控制算法的抖动更小、收敛速度更快,能够实现理想的恒力控制效果。

Aiming at the problem of force compliant control of the robotic arm interacting with the environment during operation,an adaptive impedance control algorithm based on Kalman filtering was proposed to complete the stable control of constant force in the case of unknown environmental position and environmental stiffness.Firstly,the impedance control model of the robot arm and environmental contact was constructed,and the relationship between force and position was established.Then,the adaptive impedance control model was established,and the environmental position and environmental stiffness of the robot arm contact were estimated according to the robot arm force tracking error and position error,and the control position of the robot arm was updated.The impedance control algorithm was discretized,and the impedance control system was used to Kalman filter.Finally,the stability and effectiveness of the adaptive Kalman filter impedance model were verified by simulation,and the tracking experiments were conducted by the robot to verify the effect of constant force tracking under different environmental parameters.The results show that the adaptive impedance control algorithm based on Kalman filtering can achieve the force tracking error within 5%,and the convergence speed can increase by 61.5%in the inclined environment,with less jitter and faster convergence speed compared with the traditional impedance control,which can achieve the ideal constant force control effect.