基于模型预测控制的轮腿式自平衡机器人运动控制方法  

Motion control method of wheeled-legged self-balancing robots based on model predictive control

作　　者：叶梁杰 杨忠[1] 卓浩泽 许诺 眭耀宇 何乃峰 YE Liangjie;YANG Zhong;ZHUO Haoze;XU Nuo;SUI Yaoyu;HE Naifeng(College of Automation,Nanjing University of Aeronautics and Astronautics,Nanjing 211106,China;Electric Power Research Institute of Guangxi Power Grid Co.,Ltd,Nanning 530001,China;Dongguan Power Supply Bureau of Guangdong Power Grid Co.,Ltd.,Dongguan 523000,China)

机构地区：[1]南京航空航天大学自动化学院,江苏南京211106 [2]广西电网有限责任公司电力科学研究院,广西南宁530001 [3]广东电网有限责任公司东莞供电局,广东东莞523000

出　　处：《应用科技》2024年第5期134-140,166,共8页Applied Science and Technology

基　　金：广西电网公司2023年科技创新项目(GXKJXM20230169);贵州省科技计划项目(黔科合支撑[2020]2Y044号)。

摘　　要：对于五连杆结构的轮腿式自平衡机器人,传统的倒立摆建模方法未考虑航向角等自由度,控制性能较差。对此提出一种新的建模与控制方法。首先,通过虚拟模型控制和辅助控制器实现机器人部分自由度的控制,同时对系统进行解耦,建立较为精简的整体运动模型。其次,因传统线性二次型调节器控制器无法解决腿部运动范围受限问题,设计模型预测控制器以预测未来周期内的运动状态,优化求解约束范围内的控制量,控制器输出通过虚拟模型控制映射为关节电机力矩,从而实现机器人的运动控制。最后,对轮腿式机器人进行仿真实验。仿真结果表明,该方法能够较好地跟踪控制指令的效果,具有较好的抗干扰能力。A wheeled-legged self-balancing robot with a five-link structure in its legs has poor control performance due to that traditional modeling method of inverted pendulum does not take account of heading.This paper proposes a new modeling and control approach to solve this problem.Firstly,by employing the virtual model control and auxiliary controllers,control of the robot’s partial degrees of freedom is achieved while decoupling the system to establish a simplified overall motion model.Secondly,due to the limitation of traditional linear quadratic regulator controllers in addressing the restricted range of leg motion,a model predictive controller is designed to predict the motion state within future cycles,optimizing control inputs within the constraint range.The controller output is mapped into a joint motor torque through virtual model control to achieve robot motion control.Finally,simulation experiments on the wheeledlegged robot demonstrate that this method effectively tracks control commands and exhibits good disturbance rejection capabilities.

关 键 词：轮腿式机器人 移动机器人 五连杆结构 运动控制 模型预测控制 虚拟模型控制 平衡控制 运动学与动力学 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]