Research On Path Planning Method For Collaboration Data Acquisition Task Using Multiple Autonomous Underwater Vehicles

As multiple autonomous underwater vehicles(MAUV)system can offer spatial distribution,efficiency for operation,flexibility,robustness and other advantages,it is more widely applied in the process of marine development in recent decades.Path planning is indispensable for MAUV accomplishing the task in the dynamic and complex underwater environment.Not only the autonomy for single AUV,but also the appropriate collaboration strategy between the MAUV members is required,which bring great difficulty for path planning.For the mission of underwater sensor data acquisition,this paper studies the path planning method for MAUV.To improve the autonomy and adaptive ability of the MAUV,this paper presents a behavior based path planning method.The multiple objective path planning(MOPP)model is the key content of this strategy,which includes three different basic behaviors,the corresponding behavioral objective functions and the global objective function for behavioral fusion.The three basic behaviors are defined as energy saving behavior,safety behavior and cooperative behavior among members.Then the ring function is utilized to construct the behavior objective functions,which correspond to the three basic behaviors.The decision space will be discrete to ensure the high efficiency of the objective function calculation.During the behavioral fusion(constructed the global objective function),the order weighting average operator is applied to assign the weight coefficients to each behavior objective function based on the external environment changes in real time.So the adaptability of the path planning method for the environment will be enhanced.Finally,the path planning method based on MOPP model is simulated in the MATLAB environment for different underwater environment.And the results are compared with the method without the use of ordered weighted average operator.The ultimate results show that the three behaviors defined in this paper can meet the requirements for data acquisition task.What is more,the proposed method can adjust the weight coefficients dynamically for behavior fusion in real time.So the proposed path planning method can meet the adaptability for underwater environment and the state of the autonomous underwater vehicle.