Research On Visual Simulation Platform Design And Motion Control Of Bionic Robot Fish

With the ocean power strategy proposition,underwater robots have become a crucial tool for bolstering the growth of marine resources and environmental protection.The bionic robot fish is an underwater automaton that mimics the swimming technique of fish.Due to its low noise,high flexibility,and high propulsion efficiency,it has attracted the attention of many researchers.However,the unpredictability of the marine environment makes testing autonomous bionic fish extremely difficult,and there are significant safety risks.Research on a dynamic and visualized simulation platform for underwater robots can effectively reduce test costs and risks and enhance the capacity to generate test data in extreme environments.Consequently,using the robotic fish resembling a boxfish as the research object,this paper investigated the design of the dynamic simulation platform for the bionic robotic fish,the dynamic modeling of the bionic robotic fish,and the trajectory tracking control of the bionic robotic fish,and conducted simulations and experiments to validate the relevant theory.The main research contents are as follows:Firstly,a method based on a development plan for the design and implementation of a visual simulation platform for bionic robot fish in Unreal Engine is proposed in order to conduct low-cost research on underwater bionic robotic fish.The existing simulation platform has issues such as a simple design of the simulation environment and a lack of the ability to construct large and complex scenes.This solution integrates Unreal Engine,Air Sim,and DASP technologies in virtual reality technology to provide a new solution for the development and testing of bionic robotic fish living in the ocean.In addition,it describes the simulation platform’s design,construction,and 3D modeling of robotic fish.Lastly,the simulation test is conducted on the simulation platform,and it is evident that the simulation platform has clear advantages in terms of interactivity,authenticity,visualization,and simulation effects.Secondly,to accurately analyze the motion characteristics of the bionic robotic fish,a dynamical model of the boxfish-like robotic fish’s two-dimensional plane movement was developed.According to the law of conservation of momentum and Lighthill’s theory of slender bodies,the kinematics,dynamics equations,and influence of the fluid force of the boxfish robotic fish were deduced and analyzed in detail,and then the dynamics model was developed.In addition,the simulation analysis of the model,based on the product parameters of the actual robotic fish,demonstrates the model’s viability and efficacy.The influence of these parameters on the trajectory of the robotic fish’s center of mass is summarized to provide data support for the subsequent design of the controller.Finally,a novel trajectory-tracking control strategy based on DTWPSO-MPC is proposed to realize the trajectory-tracking control of the bionic robotic fish’s autonomous swimming.Since the established dynamic model of the bionic autonomous fish includes oscillation items that are difficult to control,the original model is optimized using the averaging method,and an averaged model is obtained that makes it simple to design the controller.On this basis,a trajectory tracking controller is devised,and the controller’s weighting matrix parameters are optimized by combining the Particle Swarm Optimization(PSO)and Dynamic Time Regulation(DTW)algorithms.By simulating the tracking of three distinct curve trajectories and comparing the results to the trajectory tracking effect of typical controllers such as the Linear Quadratic Regulator(LQR),the controller’s efficacy is confirmed.The actual verification of the control effect is conducted via the pool experiment,which demonstrates the method’s excellent applicability in the real-world application setting and provides a method reference for the precise control and trajectory tracking of other marine robots.