Formation Control Of Multiple Biomimetic Robotic Fish

The research on marine ecological environment and the exploitation and utilization of resources are the most effective and feasible ways for human beings to maintain their own survival and development and thus expand their living space.Hence,the ocean will become the focus of humankind's attention in the future.The development and exploration of marine resources requires a large amount of apparatuses,and existing underwater transportation tools are unable to satisfy the demand for ocean exploration and development.Therefore,with high efficiency,flexibility and concealed performance,the underwater biomimetic robot will become an important platform during the exploration of the ocean.In future,the underwater biomimetic robot will be engaged in important tasks such as underwater biological observation,underwater work exploration,and maritime military reconnaissance,in the complex underwater environment.Compared to a single underwater biomimetic robot,the multiple underwater biomimetic robot system can accomplish more complex tasks.Thus it is important to develop efficient and stable formation control methods for a underwater biomimetic multi-robot system.Additionally the complexity of the underwater working environment and the particular propulsion mechanism of the biomimetic robotic fish,make the formation control of biomimetic robotic fish a challenging research topic.This paper focuses on the research of multiple biomimetic robotic fish formation control and obstacle avoidance issues.The main contents are as follows:(1)The formation problem of multi-robotic fish,in which all followers take a leader as reference point and track it to form formation,is first taken into consideration.For this purpose,a distributed algorithm is proposed..Firstly,based on the second-order consensus algorithm,a distributed algorithm for each followers of biomimetic robotic fish estimating the leader pose is given.Secondly,the formation description of the multiple biomimetic robotic fish formation with leader as the reference point is proposed.Moreover,according to the formation requirements,each biomimetic robotic fish determines its expected pose in the formation with the estimated reference point information.Thirdly,the biomimetic robotic fish of followers,at the speed error and angular velocity error,and the estimated leader pose information as input,the fuzzy controller is used to determine its speed gear and direction gear to realize formation and maintenance of formations.And Lyapunov function is used to prove the stability of the formation algorithm.The simulation and experimental results show that the proposed control method of distributed formation is effective.(2)Based on the second-order consensus algorithm for the formation control of multiple robotic fish,the formation control method is studied when taking the Central Pattern Generator(CPG)network as the motion controller of biomimetic robotic fish.A distributed algorithm for each biomimetic robotic fish is first estimated according to the leader pose,and the expected pose of each biomimetic robotic fish in the formation is then determined to meet the formation requirements.The expected pose of followers is selected as both a reference input and a feedback input for the CPG control network through the parameter transition layer,while PID controller is used for generating feedback signals based on velocity/angular velocity erroras.Each input parameter generates a rhythm signal in real time through the CPG network oscillator,which can accurately adjust the position of the biomimetic robotic fish on-line.The experimental results show that the biomimetic robotic fish taking on CPG control network has strongger adaptability than that in(1)when parameters are changed suddenly.Therefore,it is possible to achieve precise control of the formation.(3)The study considers the issue of multiple biomimetic robotic fish formation control in the presence of static and dynamic obstacles in the environment.A distributed formation control scheme is proposed for collision avoidance among biomimetic robotic fish and for obstacles avoidance.Firstly,using the artificial potential field method,the multiple biomimetic robotic fish formation maintenance algorithm is designed to ensure the information interaction,and then making a fixed distance between adjacent biomimetic robotic fish.Secondly,the biomimetic robotic fish can avoid obstacles by designing repulsive field and attracting field around obstacles.When the leader robotic fish tracks the target point,the follower robotic fish uses the second-order consensus algorithm to perform state matching by interacting with the information of neighboring robotic fish.Finally,simulation analysis and physical experiment of biomimetic robotic fish are carried out to illustrate the practicability of the proposed formation algorithm.