基于强化学习的切换系统综合性能优化设计

Reinforce learning based design for overall performance optimization of switched linear systems

研究连续时间切换自治线性系统综合性能优化,基于分阶段设计思路计算其最小(状态模)超调、在超调约束下的最小调节时间、在过渡过程约束下的最快指数收敛速度.借鉴强化学习方法,通过以模压缩路径的策略空间进行扩展获得有限步结束的迭代优化算法.数值仿真验证了所提算法的有效性.

This work addresses the optimization of overall performance for switched linear autonomous systems,and computes piece by piece the minimum(state norm)overshoot,the minimum settling time under the overshoot constraint,and the largest exponential convergence rate under the transient performance constraints.Utilizing the reinforce-learning approach,we obtain finite-time-stopping iterations based on extending the police space.A numerical example is presented to validate the effectiveness of these proposed algorithms.