非对合系统的鲁棒控制及在非完整机器人中的应用

This project studied the robust control of non-involutive systems with applications to nonholonomic robots. Aimed at the high nonlinearities,we first proposed a high-order sliding mode control approach to the non-involutive systems.Considering the in-matching uncertainties,we developed an explicit back-stepping design algorithm and applied to the design of high-order sliding mode controller under in-matching conditions. Besides,we also developed learning control algorithm based on the relative degree of nonlinear systems and fuzzy control algorithm based on genetic approach. The nonlinear differential geometric control principle was also used to design robust and adaptive controllers. Finally, we applied algorithms developed in this project to mobile robot systems, and performed numerical simulations and on-time implementations to show the efficiency. This project not only.palyed an important role in the control of complex nonlinear sytems, but also.provided a platform for the development of high technology products such as painting.robot in container production,autonomous cleaning robot and auto-driving systems.

以非线性算子半群理论及微分几何为主要工具研究非对合系统能控能观的充分条件和仿射非对合系统的时变非线性变换及实现方法；研究高阶变分法及伪线性化方法在非对合系统近似中的应用；采用学习控制及滑模控制等技术建立基于重构模型的鲁棒控制器设计方法；并将其应用于非完整机器人系统，实现有关算法。为非完整机器人的实际应用奠定基础。