Reinforcement Learning-Based Asynchronous Localization of USNs

Abstract

In this chapter, an autonomous underwater vehicle (AUV) aided localization issue is studied under the constraints of asynchronous time clock, stratification effect and node mobility. Particularly, an asynchronous localization protocol is constructed and then the localization problem is built to minimize the sum of all measurement errors. To solve this localization problem, we propose a reinforcement learning (RL) based localization algorithm to locate the positions of AUVs, active and passive sensor nodes. It is noted that, the proposed localization algorithm employs two neural networks to approximate the increment policy and value function, and more importantly, it is much preferable for nonsmooth and nonconvex underwater localization problem due to its insensitivity to the local optimal. Besides that, the performance analyses of proposed algorithm are given. Finally, simulation and experimental results show that the localization performance in this chapter can be significantly improved as compared with the other works.