Abstract
The formation of Autonomous Underwater Vehicle (AUV) is a promising technology. This requires multiple AUVs interact information to maintain the desired distribution of formation. At present, the formation control of AUVs has a wide range of application prospects, such as underwater environment exploration and protection. However, the formation control of AUVs becomes very complicated and difficult due to stochastic interferences in the underwater environment. Considering this situation, this paper is to study the formation problem of AUVs in underwater environment with multiple interferences. This paper formulates Multivariate Probabilistic Collocation Method (MPCM) to deal with the interferences acting on AUVs. In addition, the formation control of AUVs is also accomplished by means of reinforcement learning (RL). Particularly, the above controller can overcome multiple unknown interferences of AUVs, instead of just one specific interference, and improves the practicality of the algorithm. Finally, the effectiveness of the formation controller proposed by the simulation results is given. The results show that all the errors meet our expectations.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have