基于三角剖分的小脑模型在增强学习中的应用

研究了一种基于三角剖分的小脑模型的增强学习控制器设计方法，并应用于机器人足球中单机器人截球的学习控制中。该方法通过在Markov决策过程状态空间中引入基于单纯形的库恩三角化，实现基于三角剖分的线性值函数逼近，从而有效提高了增强学习控制器对连续状态空间马氏决策问题的泛化性能。针对机器人截球学习控制的仿真研究表明，采用基于三角剖分的小脑模型进行值函数逼近的增强学习控制器能够获得优于已有基于均匀编码的小脑模型方法的学习效率和泛化性能。

CMAC application using triangulation in reinforcement learning

A reinforcement learning controller based on CMAC neural networks using triangulation was studied and applied to the learning control of intercepting a ball in the RoboCup.By utilizing Kuhn triangulation based on simplex interpolation in the continuous state space of Markov Decision Processes(MDPs),the value functions of MDPs were approximated with linear triangulation so that the generalization ability of the CMAC-based reinforcement learning controller could be improved.Simulation results on the learning ...