摘要
针对两轮自平衡小车的平衡控制和抗干扰问题,提出基于支持向量回归机器学习方法的平衡控制器方法。分析了两轮小车的力学模型,将状态向量和控制向量运用到非线性支持向量回归算法(ε-SVR),利用LIBSVM软件对所采集样品进行训练,获得高精度的SVR模型,其中样品数据由搭建的传统闭环串级PID控制器运行得到。经过Matlab仿真测试,该模型控制器的平衡响应优于PID控制和极点配置法,最后制作了小型样车,验证了SVR模型在动态平衡和自抗干扰下存在相对优势。
A balance controller method based on support vector regression(SVR)machine learning method is proposed to address the balance control and anti-interference issues of the two-wheeled self-balancing trolley.The mechanical model of the two-wheeled trolley is analyzed,and the state vector and control vector are applied to the nonlinear support vector regression algorithm(ε-SVR).The LIBSVM is used to train the collected samples and obtain a high-precision SVR model,the sample data of which is obtained by running an established traditional closed-loop cascade PID controller.After Matlab simulation testing,the balance response of the model controller is superior to PID control and pole configuration method.Finally,a small sample trolley was made to verify the relative advantages of the SVR model in dynamic balance and self anti-interference.
作者
李加定
余光正
缪文南
孙小广
LI Jiading;YU Guangzheng;MIAO Wennan;SUN Xiaoguang(School of Electronic&Information Engineering,Guangzhou City University of Technology,Guangzhou 510800,China;School of Physics and Optoelectronics,South China University of Technology,Guangzhou 510610,China)
出处
《现代电子技术》
北大核心
2024年第11期125-130,共6页
Modern Electronics Technique
基金
国家自然科学基金面上项目(12074129)
广东省普通高校重点领域专项(2022ZDZX1041)。
