汽车工程 ›› 2019, Vol. 41 ›› Issue (5): 522-529.doi: 10.19562/j.chinasae.qcgc.2019.05.007

• • 上一篇    下一篇

无人驾驶机器人机械腿模糊监督控制*

王和荣, 陈刚   

  1. 南京理工大学机械工程学院,南京 210094
  • 收稿日期:2018-05-29 发布日期:2019-06-05
  • 通讯作者: 陈刚,副教授,博士,E-mail:gang0418@163.com
  • 基金资助:
    国家自然科学基金(51675281);江苏省六大人才高峰计划项目(2015-JXQC-003);中央高校基本科研业务费专项资金项目(30918011101);江苏省研究生科研与实践创新计划项目(KYCX18_0395)资助

Fuzzy Supervisory Control of Mechanical Legs of Unmanned Robots

Wang Herong, Chen Gang   

  1. School of Mechanical Engineering, Nanjing University of Science and Technology, Nanjing 210094
  • Received:2018-05-29 Published:2019-06-05

摘要: 为实现对无人驾驶机器人机械腿运动的精确控制,提出了一种模糊监督控制方法。通过对驾驶机器人机械腿操纵自动挡汽车油门/制动踏板的运动分析,描述了机械腿各杆件的运动学关系,并建立了机械腿的拉格朗日动力学模型。在此基础上,设计了一种模糊监督控制器,并通过Lyapunov稳定性分析原理,验证了跟踪误差的收敛性,保证了机械腿对位移跟踪的稳定性。模糊监督控制器以机械腿的位移跟踪误差及误差变化率为输入,位移跟踪过程中,实时监测跟踪误差的变化趋势,当误差不超过给定值时,模糊控制器单独作用,当误差超出给定值,采用模糊监督控制器。最后,设计了一种油门/制动机械腿切换控制器,搭建了油门/制动机械腿车速跟踪仿真模型,仿真结果与实车试验数据比较,验证了提出方法的有效性。

关键词: 无人驾驶机器人, 动力学模型, 模糊监督控制, 稳定性

Abstract: To achieve precise control of the leg of a robot driver, a fuzzy supervisory control method is proposed. By kinematic analysis of the throttle /brake pedal of a driving robot manipulated by the mechanical leg, the kinematic relationship of each leg of mechanical legs is described and the Lagrange dynamic model of mechanical legs is established. On this basis, a fuzzy supervisory controller is designed. The stability of tracking error is verified by the principle of Lyapunov stability analysis, and the stability of displacement tracking is ensured. The fuzzy controller takes the displacement tracking error and the rate of error change as input. In the process of displacement tracking, the change trend of tracking error is monitored at all times. When the error does not exceed the given value, the fuzzy controller acts alone; when the error exceeds the given value, the fuzzy supervisory controller is used. Finally, a throttle/brake mechanical leg switch controller is designed, and the speed tracking simulation model of the accelerator/brake leg is built. By comparing the simulation results with the experimental data, the effectiveness of the proposed method is verified.

Key words: unmanned robot, dynamic model, fuzzy supervisory control, stability