Fuzzy C-means method for representation policy iteration in reinforcement learning | IEEE Conference Publication | IEEE Xplore