Multi-agent Cooperative Control And Its Application In Pipeline Leakage Source Tracking

With the development of artificial intelligence technology,multi-agent system,which is an important branch of artificial intelligence technology,has attracted the attention of a large number of domestic and foreign researchers.However,the application of multi-agent system in some fields has become a hot research topic.Based on the practical application of multi-agent system in chemical environment,this paper studies the application of cooperative control of multi-agent system in pipeline leak source tracking.Firstly,the multi-agent distributed architecture,which is more suitable for this research,is selected.The internal structure of the multi-agent and its the specific workflow are described in detail.Meanwhile,the consistency theory of the multi-agent system is put forward.It provides a theoretical basis for the simulation of the consistency problem in the next formation control process.Secondly,in the research of multi-agent formation tracking control,this paper studies on formation of multi-agent formation by using l-? rigid control strategy and leader-follower method.The research shows that after the formation of the multi-agent formation,the motion track meets the consistency requirement,and when the detected leakage point is changed,the agent can automatically change the leader to perform real-time formation tracking.The obstacle avoidance control in formation tracking task is also studied.Two kinds of obstacle avoidance methods,hybrid autonomous obstacle avoidance and cooperative obstacle avoidance,are designed and discussed respectively.Lastly,the concentration distribution of the leakage gas in the leakage area is simulated based on the gaussian plume model which is more suitable for continuous point source diffusion,and the tracking and positioning of the multi-agent to the leakagesource is accomplished by using the concentration gradient method and the upwind search method.The simulation results show that the gaussian plume model can simulate the plume distribution of the leaking gas and multi-agent robots can track and locate the leaking point by applying the concentration gradient method and the upwind search method.