Integrated Localization and Tracking for AUV With Model Uncertainties via Scalable Sampling-Based Reinforcement Learning Approach
نویسندگان
چکیده
This article studies the joint localization and tracking issue for autonomous underwater vehicle (AUV), with constraints of asynchronous time clock in cyberchannels model uncertainty physical channels. More specifically, we develop a reinforcement learning (RL)-based algorithm to localize position AUV, where AUV is not required be well synchronized real time. Based on estimated position, scalable sampling strategy called multivariate probabilistic collocation method orthogonal fractional factorial design (M-PCM-OFFD) employed evaluate time-varying uncertain parameters AUV. After that, an RL-based controller designed drive desired target point. Besides performance analyses integration solution are also presented. Of note, advantages our highlighted as: 1) can avoid local optimal traditional least-square methods; 2) M-PCM-OFFD-based address reduce computational cost; 3) communication energy consumption. Finally, simulation experiment demonstrate that proposed effectively eliminate impact clock, more importantly, M-PCM-OFFD find accurate optimization solutions limited costs.
منابع مشابه
Hybrid coordination of reinforcement learning-based behaviors for AUV control
This paper proposes a Hybrid Coordination method for Behavior-based Control Architectures. The hybrid method takes in advantages of the robustness and modularity in competitive approaches as well as optimized trajectories in cooperative ones. This paper shows the feasibility of this hybrid method with a 3D-navigation application to an Autonomous Underwater Vehicle (AUV). The behaviors were lear...
متن کامل(More) Efficient Reinforcement Learning via Posterior Sampling
Most provably-efficient reinforcement learning algorithms introduce optimism about poorly-understood states and actions to encourage exploration. We study an alternative approach for efficient exploration: posterior sampling for reinforcement learning (PSRL). This algorithm proceeds in repeated episodes of known duration. At the start of each episode, PSRL updates a prior distribution over Mark...
متن کاملstatistical inference via empirical bayes approach for stationary and dynamic contingency tables
چکیده ندارد.
15 صفحه اولModel-Based Reinforcement Learning for Partially Observable Games with Sampling-Based State Estimation
Games constitute a challenging domain of reinforcement learning (RL) for acquiring strategies because many of them include multiple players and many unobservable variables in a large state space. The difficulty of solving such realistic multiagent problems with partial observability arises mainly from the fact that the computational cost for the estimation and prediction in the whole state spac...
متن کاملSmarter Sampling in Model-Based Bayesian Reinforcement Learning
Bayesian reinforcement learning (RL) is aimed at making more efficient use of data samples, but typically uses significantly more computation. For discrete Markov Decision Processes, a typical approach to Bayesian RL is to sample a set of models from an underlying distribution, and compute value functions for each, e.g. using dynamic programming. This makes the computation cost per sampled mode...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: IEEE transactions on systems, man, and cybernetics
سال: 2022
ISSN: ['1083-4427', '1558-2426']
DOI: https://doi.org/10.1109/tsmc.2021.3129534