Heuristic Dynamic Programming With Internal Goal Representation
OSTP Journal Article
University of Rhode Island Kingston United States
Pagination or Media Count:
In this paper, we analyze an internal goal structure based on heuristic dynamic programming, named GrHDP, to tackle the 2-D maze navigation problem. Classical reinforcement learning approaches have been introduced to solve this problem in literature, yet no intermediate reward has been assigned before reaching the final goal. In this paper, we integrated one additional network, namely goal network, into the traditional heuristic dynamic programming HDPdesign to provide the internal rewardgoal representation. The architecture of our proposed approach is presented, followed by the simulation of 2-D maze navigation 1010 problem. For fair comparison, we conduct the same simulation environment settings for the traditional HDP approach. Simulation results show that our proposed GrHDP can obtain faster convergent speed with respect to the sum of square error, and also achieve lower error eventually.