自适应卡尔曼滤波器在机器人控制中的应用-中国科技论文在线.PDF

中国科技论文在线 136 西安理工大学学报Journalof Xi. an University of Technology(2007) Vol. 23 No.2 : 1006-4710(2007)02-0136-04 自适应卡尔曼滤波器在机器人控制中的应用 辛菁, 刘丁, 班建安 (西安理工大学自动化与信息工程学院, 陕西西安710048) : 机器人控制系统在实际工作中不可避免地要受到随机噪声的影响, 当噪声的统计特性已知 时, 可以考虑采用常规Kalman 滤波以抑制随机噪声对控制性能的影响;但当噪声的统计特性不完 全已知时, 常规Kalman 滤波的滤波性能会下降甚至会引起发散 根据机器人的动态特性, 设计了 一个自适应Kalman 滤波器, 并对该滤波器应用于机器人控制系统进行了仿真实验研究 仿真结 果表明,所设计的滤波器能够较好地抑制方差和均值未知的测量噪声对机器人控制系统的影响,控 制系统的动态性能得到了较大的改善 : 随机噪声; 自适应Kalman 滤波; 机器人控制 : TP273 : A Application Research of Adaptive Kalman Filter in Robot Control XIN Jing, LIU Ding, BAN Jian-an (Faculty of AutomationandInformation Engineering, Xi. anUniversity of Technology,Xi. an 710048,China) Abstract: Robot control system is influenced inevitably by random noise during its running. When statisticalproperty of random noise is known, generalkalman filter can be adopted to hold back effect of measurement noise on control performance. When statistical property of random noise is not all known completely, filter performance would drop and even result in divergence. Based on the dynamics nonlinearities of robot manipulator, a adaptive kalman filter is designed and applied to robot controlsystem in simulation research. Simulation results show that the de- signed filter can holdback effect of measurement noise with unknown varianceandmean on robot control system,and that the dynamic performance of controlsystem is improved. Key words: random noise; adaptive kalman filter; robot control 机器人是一个复杂的多输入多输出的非线性系 器, 如维纳滤波和卡尔曼滤波 Kalman 滤波是R E 统, 在众多的机器人控制算法中, PD 和PID 是最简 Kalman 于 1960 年提出的, 它是从已被提取信号有 单的控制算法, 目前绝大多数工业机器人都采用这 关的观测量中通过算法估计出所需信号的一种滤波 [1,2] 类控制算法 。实验证明, 即使像机器人这类强