基于MPC的四足机器人斜面稳定控制研究  

Research on Slope Stability Control of Quadruped Robert Based on MPC

作　　者：赵宇镇 杨旗[1] Zhao Yuzhen;Yang Qi

机构地区：[1]沈阳理工大学机械工程学院,辽宁沈阳110159

出　　处：《一重技术》2024年第6期71-74,65,共5页CFHI Technology

基　　金：辽宁省研究生教育教学改革研究资助项目“校企深度融合智能制造研究生导师团队建设探索与实践”(项目编号:LNYJG2023077)。

摘　　要：针对使用模型预测控制算法在斜面上运动的四足机器人遇到外力等因素干扰时运动不稳定问题,利用以惯性测量单元(IMU)测得机器人的运动姿态参数对MPC控制器预测步长进行优化,增强控制调节实时性,将模糊控制理论引入误差补偿,对误差变化程度进行划分,利用误差补偿系数实时补偿模型预测偏差,增强四足机器人的运动稳定性,通过MATLAB仿真,证明在外力干扰下,改进的机器人能够在不同斜面稳定运行,验证改进方法的有效性和正确性。In view of the problems of motion instability of quadruped robot moving on an slope surface when it encounters external force and other factors that has been found during application of model predictive control algorithm,this paper proposes a method of optimizing the prediction step length of MPC controller by using the motion posture parameters of the robot measured by inertial measurement unit(IMU),improving the capacity of real-time control adjustment,introducing the fuzzy control theory into error compensation to separate the degree of error variation,and using the error compensation coefficient to realize real-time compensation of the model prediction deviation,so as to improve motion stability of the quadruped robot.Through MATLAB simulation,it is proved that after improvement,the quadruped robot is able to make stable motions on different slope surfaces,and thus verifies the effectiveness and correctness of the improved method.

关 键 词：模型预测控制 外界因素 模糊控制 MATLAB 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置] TP391.9[自动化与计算机技术—控制科学与工程]