Theoretic and Experimental Research on The Nonlinear Control of Nonlinearly Actuated Systems and The Steering of Mobile Robots

本研究的目的是开发实用的非线性控制方法，用于非线性驱动系统和移动机器人的转向。并开发了相应的控制系统包。主要研究成果如下：1.针对一般的移动机器人--链式系统，提出了一种切换控制算法。该方法具有全局指数收敛的特点，可以应用于空间约束和转角约束等物理约束条件下的移动机器人转向控制。2.基于上述算法开发了一个能够处理车库约束和转角约束下的停车任务的实际自动停车系统。它的高适用性已被大量的实验所验证。此外，为了更有效地完成停车任务，通过引入辅助停车目的地，建立了各种停车算法。3.作为…对于自动停车控制，更多的是基于跟踪单车族的思想提出了另一种方法。针对典型的非线性驱动系统--电磁轴承系统，建立了一种基于低偏置的非线性动态输出反馈控制方法。5.将这种动态输出反馈控制方法推广到具有时变转速和柔性的被控对象。6.作为对偶问题，研究了线性驱动器驱动的非线性系统的控制问题。定量地分析了直线驱动器在保证整个系统的稳定性时应达到多快的问题，并成功地实现了单轮车机器人的运动控制和旋转起重机的高速非线性控制。进一步，在H2控制器的模型降阶、电动系统的自适应控制以及H_∞控制解的新推导等方面得到了新的认识。从而达到了本研究的目的。较少

This research aims at developing practical nonlinear control methods for nonlinearly actuated systems and the steering of mobile robots. The corresponding control system packages are also developed. The obtained results are listed below :1.A switching control algorithm has been proposed for chained systems, a general class of mobile robots. This method guarantees the global exponential convergence and has a distinctive feature that it can be applied to the steering control of mobile robots with physical constraints such as space constraint and steering angle limitation.2.A realistic automatic parking system has been developed based on the above algorithm, which is able to handle parking tasks subject to garage constraint and steering angle limitation. Its high applicability has been validated by numerous experiments. Further, in order to achieve parking task more efficiently various parking algorithms have been established via the introduction of auxiliary parking destinations.3.As an … More alternate for automatic parking control, another method has been proposed based on the idea of tracking of a cycle family. Its effectiveness has also been verified by experiments.4.A nonlinear dynamic output feedback control method using low bias has been established for electromagnetic bearing systems, a typical nonlinearly actuated system. By the use of this method, the bias current is reduced dramatically.5.Moreover, this dynamic output feedback control method is extended to handle plants with time-varying rotation speed and flexibility.6.As a dual problem, the control of nonlinear systems driven by linear actuators has also been studied. A quantitative analysis has been conducted to answer the question that how fast should the linear actuator be in order to ensure the stability of the whole system.We have also succeeded in the motion control of a unicycle robot and the high speed nonlinear control of a rotary crane. Further, new knowledge have been obtained on the model reduction of H2 controllers, adaptive control of motor-driven systems and new derivation of H_∞ control solution. As a result, the objective of this research has been achieved. Less