Font Size: a A A

Formation Control With Collision Avoidance Of Nonlinear Multi-Agent Systems And Autonomous Surface Vehicles Applications

Posted on:2021-05-06Degree:MasterType:Thesis
Country:ChinaCandidate:Q ShiFull Text:PDF
GTID:2392330602489143Subject:Traffic Information Engineering & Control
Abstract/Summary:PDF Full Text Request
With the development of science and technology and requirements of production,the requirements for task scale and work efficiency have been greatly grown.People realize that using only a single autonomous individual has become increasingly difficult to deal with increasingly complex comprehensive tasks,and swarm intelligence has naturally become a technical problem that needs to be solve.Because the distributed cooperative control of multi-agent system(MASs)has both the independence and autonomy of the monomer,and the coordination and cooperation of the swarm,it is considered to be one of important research directions in the field of control theory and control engineering by researchers.In the field of marine engineering,it has also been widely used in the fleet's cooperative formation operations and maritime joint search and rescue.Undoubtedly,the research on distributed coordination formation control with collision avoidance technology of multi-agent or multi-vehicle is the most important in these applications.Although domestic and foreign researchers have made certain breakthrough results on this topic,the research results are concentrated on the theoretical research of linear systems.The controlled systems.often have unknown nonlinear dynamics in actual engineering,.such as autonomous surface vehicles(ASVs)and underwater robots.Therefore,based on related researches,this paper studies the schemes of formation control with collision avoidance of MASs with unknown nonlinear dynamics.The problem of collision avoidance between agents in the stage of formation forming is discussed in detail.Then the theoretical research was developed and successfully applied to the research of distributed formation control with collision avoidance of multiple AS Vs.The main works of the paper are displayed as follows:Firstly,this paper proposes the formation control schemes with collision avoidance of MASs with unknown nonlinear dynamics from the perspective of theoretical control.Aiming at the negative impact of unknown nonlinear dynamics on system control performance and multi-agent collision avoidance control,neural network(NN)technologies are used to approximate the unknown nonlinear dynamics in the system online in this paper.Based on the artificial potential field(APF)theory,the adaptive NN formation control algorithms with collision avoidance of nonlinear MASs are designed.Secondly,this paper proposes the distributed formation control schemes with collision avoidance of AS Vs with unknown nonlinear dynamics from the perspective of practical engineering.Aiming at the problems that the centralized formation control structure is limited by the influence of communication bandwidth and the excessive dependence of the monomer on the central centralized control unit.Combining the graph-based methods and the leader-follower formation structure,the distributed formation control strategies with collision avoidance control of AS Vs based on APF theory are designed in this paper.In view of the characteristics of coupling between the controller input and state output in the control model,the above MASs formation control schemes with collision avoidance cannot be directly used in the control problem of ASVs.The coordinate transformation methods are adopted to decouple the coupling characteristics of the systems.Based on NN technologies,the unknown nonlinear dynamics associated with the internal unmodeled dynamics,external environmental disturbances and time-varying hydrodynamics are approximated online.The distributed adaptive NN formation control algorithms with collision avoidance of AS Vs with unknown nonlinear dynamics are designed.In this paper,Matlab software is used to simulate the theoretical algorithm,and the simulation results verify the effectiveness of the proposed algorithm to achieve nonlinear multi-agent system and multi-ASV formation collision avoidance control.Based on the research results of control theory of MASs formation with collision avoidance and the application of control engineering research of AS Vs formation with collision avoidance,from the theoretical perspective,it is helpful to solve the cooperative control of an autonomous swarm system with unknown nonlinear dynamics,and then promote the transformation of relevant theories into practical applications.From the practical perspective,this research lays an important foundation for improving the maritime operations safety of multiple marine vehicles and achieving efficient shipping.
Keywords/Search Tags:Nonlinear Multi-agent Systems, Autonomous Surface Vehicles, Formation Control with Collision Avoidance, Neural Network, Artificial Potential Field Methods
PDF Full Text Request
Related items