CAREER: Guidance and Control of Fish Shoals using Bio-Mimetic Robots

许多多车辆团队和移动的传感器网络的控制策略直接源于对动物群体集体行为的数学理解。鱼群、鸟类、群居有蹄类动物的聚集，以及人群中的交通流模式，都激发了协调机器人系统的算法。 然而，迄今为止，工程与自然之间的关系在很大程度上是单向的。该学院早期职业发展（CAREER）研究计划旨在通过仿生机器鱼对鱼群的工程指导和控制进行多维研究，从而实现工程与自然之间的“闭环”。 这项工作将提高目前对复杂多智能体动力系统的理解，对多车辆机器人技术，动物行为科学，水产养殖工程和鱼类保护方法具有潜在的影响。这个CAREER计划将开发一个全面的动力系统框架，用于研究由活体和仿生机器鱼组成的鱼群中的领导力。机器人领导执法的不同机制，包括复制活浅滩领导人的关键特征和使用受控刺激，如声音，水流，光和化学物质，鱼类种群，将进行调查。行为模型和复杂网络的分析和控制的数学方法将先进的理解和控制鱼群的动态。鱼群引导问题将被表述为加权时变邻近网络上的钉扎控制问题。 微型机器鱼将使用基于新兴智能材料（如离子聚合物金属复合材料）的多功能传感器和致动器进行开发。拟议的研究将与当地大型动物水族馆的创新教育经验相结合，为小学，初中和高中学生，包括那些代表性不足的群体。这种经验的目的是激励年轻人追求工程和科学事业。

Many control strategies for multi-vehicle teams and mobile sensor networks directly stem from a mathematical understanding of the collective behavior of animal groups. Schooling of fish, flocking of birds, herding of social ungulates, and patterns of traffic flow in human crowds inspire algorithms for coordinating robotic systems. However, the relationship between engineering and nature has been largely one-directional until now. This Faculty Early Career Development (CAREER) research program aspires to "close the loop" between engineering and nature by a multidimensional study of engineered guidance and control of fish shoals through bio-mimetic robotic fish. This work will improve the current understanding of complex multi-agent dynamical systems, with potential impact on multi-vehicle robotics, animal behavior science, aquacultural engineering, and fish protection methods.This CAREER program will develop a comprehensive dynamical systems framework for studying leadership effectiveness in fish shoals comprising of live and bio-mimetic robotic fish. Different mechanisms for robot leadership enforcement, including the replication of key traits of live shoal leaders and the use of controlled stimuli, such as sound, water flow, light, and chemicals, to the fish population, will be investigated. Behavioral models and mathematical methods for analysis and control of complex networks will be advanced to understand and control the dynamics of fish shoals. The fish shoal guidance problem will be formulated as a pinning control problem over a weighted time-varying proximity network. Miniaturized robotic fish will be developed using multifunctional sensors and actuators based on emerging smart materials, such as ionic polymer metal composites. The proposed research will be integrated with an innovative educational experience at a local large animal aquarium for elementary, middle, and high school students, including those from under-represented groups. The objective of this experience is to inspire young minds in pursuing careers in engineering and sciences.