基于群智能的仿生机器鱼集群未知水域游动控制

仿生机器鱼集群控制是仿生领域新的研究热点，然而要使机器鱼像自然界鱼群那样在未知水域高效、机动地聚集和分散依然面临诸多难题。本项目拟针对仿生机器鱼集群在未知水域的动力学建模、游动控制、队形控制问题开展深入研究，主要内容包括：首先，考虑未知动态环境下可能存在水流、波浪、邻近机器鱼、障碍物、船只经过等影响因素，建立仿生机器鱼的动力学模型，解决机器鱼个体的流固耦合建模难题；其次，针对扰动情形下的机器鱼个体游动控制，设计非线性扰动观测器，提出机器鱼稳定游动的控制策略；再次，在弱通信条件和输入状态受限情况下针对多仿生机器鱼队形控制进行研究，提出基于非线性预测控制方法的一致性策略；最后，基于人工鱼群理论进行仿生机器鱼集群控制进行研究，提出仿生机器鱼群在围捕觅食和躲避天敌逃生的极端情况下的队形形成、保持和切换方法。本项目研究结果将为仿生机器鱼集群及类似多机器人系统在未知环境的导航、协作与控制提供技术支撑。

Bionic robotic fish swarm control is a new research hotspot in the field of bionics. However, it still faces many problems, such as modeling, navigation, positioning, control, etc., to make robotic fish as efficient and maneuverable as natural fish swarm in unknown waters. This proposal will explore dynamic modeling, swimming control, and formation control of bionic robotic fish swarm in unknown and dynamic environments. The main research contents are listed as follows. Firstly, considering the possible influence factors such as current, wave, adjacent robotic fish, obstacles and ship passing in unknown waters, the dynamic model of bionic robotic fish will be established based on hydrodynamics to solve the problem of fluid-solid interaction. Secondly, the control strategy of the robotic fish stable swimming will be proposed by designing a nonlinear disturbance observer. Thirdly, under the condition of weak communication and limited input state, the control strategy of the multiple bionic robotic fish formation will be presented. Finally, based on the theory of artificial fish swarm, the control of bionic robotic fish swarm will be studied, and the formation, maintenance and switching methods of bionic robotic fish swarm in the extreme situation of hunting, foraging, and escaping from natural enemies will be proposed. The results of this proposal will provide technical support for navigation, cooperation, and control of bionic robotic fish swarm and similar multi-robot systems in unknown and dynamic environments.