基于模糊神经网络的深海集矿机路径规划

探讨了深海多金属结核集矿机在作业过程中的实时局部路径规划问题，旨在解决在深海复杂特殊环境下多金属结核集矿机的自适应实时路径规划问题。采用了能实现模糊控制规则的基于强化学习方法的自学习和自调整算法来实现深海集矿机的实时运动规划，并提出了能实现模糊控制规则的基于强化学习的自学习和自调整的规划算法。设计了深海集矿机实时运动规划器结构、规划器操作过程以及相应的算法。集矿机试验样机池试试验表明该方法的有效性．可广泛应用于类似路径规划问题。

Path Planning for Deep-sea Mineral Collector Based on Fuzzy Neural Network

A real-time local path planning problem in the deep-sea multi-metal nodule collector working process is discussed. The studying purpose is to solve real-time local path planning problem of the mine collector in complicated deep-sea environment. A self-learning and self-turning algorithm is proposed based on reinforcement learning which can implement fuzzy control rules,and designed the deep-sea mine collector real-time mobile planer architecture,working process and relevant algorithm are designed. The pool experimentation proves the efficiency of the method which presented and the method can be used in similar path planning problem.