Heuristic Dynamic Programming With Internal Goal Representation

Abstract

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 (HDP)design to provide the internal reward/goal representation. The architecture of our proposed approach is presented, followed by the simulation of 2-D maze navigation (10*10) 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.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 03, 2013
Accession Number
AD1022355

Entities

People

  • Haibo He
  • Zhen Ni

Organizations

  • University of Rhode Island

Tags

Communities of Interest

  • Autonomy
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Adaptive Systems
  • Algorithms
  • Artificial Intelligence
  • Artificial Intelligence Software
  • Computational Science
  • Computer Programming
  • Dynamic Programming
  • Kalman Filters
  • Machine Learning
  • Motion Planning
  • Multiple Input Multiple Output
  • Neural Networks
  • New York
  • Recurrent Neural Networks
  • Reinforcement Learning
  • Simulations
  • Two Dimensional

Fields of Study

  • Computer science

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Computational Fluid Dynamics (CFD)
  • Operations Research

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms