一种新的机器人自适应控制方式

One New Adaptive Control Method for Robot Manipulators

摘要: 本文提出了一种新的、有效的机器人自适应控制方式，克服了其他方法由于模型不准或计算量大等所带来的一系列问题．本文首先将 Lagrange 运动方程转化为 ARMA 模型，并用虚拟噪声补偿模型误差（即由于线性化、解耦、观测不准和干扰等误差）．然后利用改进的 Kalman 自适应滤波算法在线进行参数辨识和状态估计，将获得的参数用于机器人控制系统自适应控制器的设计．最后给出了该算法的仿真结果并对此进行了讨论．

Abstract: The method of using fictitious noisy compensation technique was proposed in this paper.We derived the mathematical model of robot manipulators based on Lagrange equation of motion and introduced fictitious noisy to compensate the model errors.In addition,we discussed parameter estimation problems for generalized stochastic moving model and presented the control law by using improved Kalman filtering algorithm.Finally,the simulation results on PDP 11/23 computer were given.