项目名称: 基于有限时间切换的混合系统可控虚拟生存域研究与应用
项目编号: No.61202159
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 计算机科学学科
项目作者: 李坚强
作者单位: 深圳大学
项目金额: 25万元
中文摘要: 随着CPS、智能机器人、智能交通等复杂系统的发展,单一离散或连续方程已难以描述其特征。本项目拟采用混合系统的方法对其进行建模和分析,但是当前研究存在以下问题:(1)缺少对非理想切换瞬态的建模,(2)生存域的描述与求解比较复杂。 针对这些问题,本项目进行如下的研究:(1)提出一种含有限时间切换的混合系统建模方法,研究其非理想切换下的分段仿射和混合逻辑动态模型以及线性摄动约简;(2)求解其最大不变集的生存域及其表示方法,通过逆向计算建立瞬态过程所对应的虚拟生存域空间;(3)提出一种基于不变集的混合逻辑模型预测优化控制算法,使系统在性能最优的情况下具有较好的鲁棒性和安全性;(4)将理论结果应用于机器人系统新型优化切换策略设计。 项目从状态集的角度进行研究,把混合系统理想模型拓展到"有限时间切换"的非理想模型,对于复杂系统建模、状态集计算以及优化控制设计有着重要理论价值,并具有广阔的应用前景。
中文关键词: 混合系统;不变集;混合逻辑动态规划;有限时间切换;生存域
英文摘要: As the development of complex systems such as cyber-physical systems(CPS),robotics and inteligent traffic, the system characters can not be precisely described by simple equations. To overcome the disadvantage of hybrid systems' ideal transition model, as well as the problem of survival region complex description and solving method, this project proposes the research of finite-time transition hybrid systems and its virtual viability regions. For actual system modeling problems, this project proposes a finite-time transition hybrid system modeling method. The non-ideal transition under the piecewise affine (PWA) model and mixed logic dynamic (MLD) model as well as linear perturbation reduction are studied. Then the viability region of maximal invariant set and its expression method are proposed to solve through establishing virtual survival instant state regions by reverse calculation. Finally mixed logic model predictive control algorithm is proposed based on invariant sets; the transition sequence and conditions are designed to obtain better robustness and security. Theoretical results will be directly applied to design optimal transition strategy for robot systems. The state set of hybrid systems is studied in this project, and hybrid systems ideal model is expanded to "finite-time transition" model. This re
英文关键词: hybrid systems;invariant sets;mixed logic dynamic program;Finite-time Transition;Viability Region