基于卡尔曼滤波和LQR算法的两轮自平衡巡检机器人的设计与应用-控制理论与控制工程专业论文.docxVIP

ABST ABSTRACT PAGE PAGE III ABSTRACT Due to the complexity of the environment in substation and nuclear power station, high flexibility robot has been the research hotspot in robot field. Instead of artificial inspection, the remote control inspection robot will be the development tendency in substation and nuclear power station inspection domain. This paper designed a two-wheeled self-balanced inspection robot which can move freely on flat ground and realize remote control. The complex structure within the substation and nuclear power station requires high flexibility of the inspection vehicle, so the robot applies four-wheel driven structure. Mechanical structure and control circuit of the robot was designed in this robot. The mechanical structures include the choice of wheel and body, and the installment of batteries and sensors. The control circuits include the minimum system that STM32F103C8 as the control chip, power module, motor driven module, state detection module, communication module, the temperature detection module and camera. In order to control the robot accurately, dynamics and kinematics model of the system was established. On the choice of control algorithm, due to the two-wheeled self-balance robot system has the characteristic of multiple state variables, LQR (Linear Quadratic Regulator) control algorithm was used to control the robot system. After the analysis of the necessity of data fusion, Kalman filtering algorithm was used to fuse dip angle measured values to acquire more accurate dip angle to improve the stability of the robot system. In order to control the robot system via wireless network, PC client which was established by Microsoft NetFramework 4.0 and written by C# was designed. After related configuration, robot system could send robot status and environment information to PC client, and receive the information from PC client. Kalman experiment, system simulation experiment and system real-time control experiment were conducted