引用本文:孙宁,方勇纯,陈鹤.欠驱动惯性轮摆系统全局滑模控制[J].控制理论与应用,2016,33(5):653~661.[点击复制]
SUN Ning,FANG Yong-chun,CHEN He.Global sliding mode control of underactuated inertia wheel pendulum systems[J].Control Theory and Technology,2016,33(5):653~661.[点击复制]
欠驱动惯性轮摆系统全局滑模控制
Global sliding mode control of underactuated inertia wheel pendulum systems
摘要点击 2755  全文点击 1992  投稿时间：2015-02-11  修订日期：2016-05-22
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/CTA.2016.50135
  2016,33(5):653-661
中文关键词  欠驱动系统  惯性轮摆  滑模控制  不确定性系统
英文关键词  underactuated systems  inertia wheel pendulums  sliding mode control  uncertain systems
基金项目  国家自然科学基金(61503200, 11372144), 天津市应用基础与前沿技术研究计划(青年项目)(15JCQNJC03800); 中央高校基本科研业务费项目资助.
作者单位E-mail
孙宁 南开大学机器人与信息自动化研究所 sunn@nankai.edu.cn 
方勇纯* 南开大学机器人与信息自动化研究所 fangyc@nankai.edu.cn 
陈鹤 南开大学机器人与信息自动化研究所  
中文摘要
      针对欠驱动惯性轮摆的镇定控制问题, 本文提出了一种新型的滑模鲁棒控制策略, 可在系统受到不确定性 与外界干扰影响的情况下, 实现全局渐近镇定控制. 区别于现有方法, 本文方法无需切换, 且能将无驱动的摆杆摇起 至竖直向上位置的同时, 确保惯性轮回到初始位置. 具体而言, 首先对惯性轮摆系统的非线性模型进行非奇异坐标 变换, 将其变为类积分器形式. 随后, 根据转换后系统的形式, 构造了一种新型的滑模面; 经严格分析知, 当系统状 态处于该滑模面上时, 它们将渐近收敛于平衡点. 在此基础之上, 设计了滑模控制律以确保系统状态始终处于该滑 模面上, 以实现镇定控制. 最后, 通过仿真验证了所提控制方法的有效性与鲁棒性, 并与现有方法进行了对比.
英文摘要
      For the stabilization problem of underactuated inertia wheel pendulums, a new sliding mode robust control strategy is proposed, which can achieve global asymptotic stabilization in the presence of uncertainties and disturbances. Different from existing methods, the proposed approach needs no switching between different control laws, and it can swing up the unactuated pendulum to its upright position while making the inertia wheel return to its initial position. Specifically, some nonsingular coordinate changes are performed to transform the nonlinear model into a quasi-chain-of-integrators form. Based on the transformed model, a new sliding surface is constructed, and rigorous analysis is implemented to prove the convergence of the state variables to the equilibrium point when they are kept on the sliding surface. Based on that, a sliding mode control (SMC) law is designed to keep the state variables always staying on the constructed sliding surface, in order to realize stabilization control. We validate the effectiveness and robustness of the proposed controller and compare it with existing methods through simulation results.