基于演化博弈的智能群体控制理论及在共融机器人协作中的应用

Mechanism of swarm coordination is one of the key technologies for multi-robot cooperation which has an important application value. This proposal focuses on the theory of smart swarm control which combines multi-agent systems and the evolutionary games theory, and conducts experimental research on the cooperation of multiple underwater robots. Evolutionary game theory is introduced to the classical multi-agent system and the start swarm system will be modeled. We will analyze its structural performance, such as controllability and observability; investigate the coevolution of the dynamics of cooperation in the population and the dynamics of system; decide some decisive factors and mechanisms which affect the level of cooperation, and analyze how they affect the system dynamics. Then, we will integrate the problems of consensus, containment control and formation control; further propose some reasonable performance indices and give the mathematical description and model of the performance optimization problem; design control protocols and the strategy update rules in our proposed coevolutionary dynamics model to solve the optimization problem. Finally, we will use several biomimetic robotic fish to built an experimental platform for the formation control of multiple robotic fish, conduct experiments to verify and evaluate the effectiveness of our proposed theory and methods. The research results of this proposal will provide strong support to perfect the research on multi-robot cooperation and swarm intelligent control.

多共融机器人的协同作业技术具有巨大应用价值，群体协作控制机制与方法是其重要理论基础和关键支撑技术。本项目拟研究融合多智能体系统和合作演化动力学的智能群体控制理论，并针对多水下机器人协作展开实验研究。首先，将演化博弈理论引入经典多智能体系统模型中，提出一个融合演化博弈理论和多智能体系统动力学的智能群体系统。其次，对所建智能群体系统模型进行可控可观性等结构性能分析，并研究群体合作动力学和多智能体系统动力学的共演化的问题，确定若干影响合作水平的决定性的因素和机制，定量分析其对系统动力学的影响。接着，展开系统一致性、包围控制与编队控制等问题的研究，进一步，提出合理性能优化指标，设计控制协议和演化规则，对系统进行性能优化。最后，以仿生机器鱼为对象，搭建多机器鱼编队控制实验平台，通过实验对智能群体理论与方法的可行性与有效性进行分析与评价。本项目成果将有助于从理论上拓展多机器人协作与群体智能控制的研究。

多共融机器人的协同作业技术具有巨大应用价值，群体协作控制机制与方法是其重要理论基础和关键支撑技术。本项目拟研究融合多智能体系统和合作演化动力学的智能群体控制理论，并针对多水下机器人协作展开实验研究。首先，将演化博弈理论引入经典多智能体系统模型中，提出一个融合演化博弈理论和多智能体系统动力学的智能群体系统。其次，对所建智能群体系统模型进行性能分析，并研究群体合作动力学和多智能体系统动力学的共演化的问题。接着，展开系统一致性、包围控制与编队控制等问题的研究，进一步，提出合理性能优化指标，设计控制协议和演化规则，对系统进行性能优化。最后，以仿生机器鱼为对象，展开仿真研究和实验研究，搭建多机器鱼协作控制仿真平台和实验平台，通过实验对智能群体理论与方法的可行性与有效性进行分析与评价。本项目成果将有助于从理论上拓展多机器人协作与群体智能控制的研究。