Deep Reinforcement Learning with Heuristic Corrections for UGV Navigation.
Changyun WeiYajun LiYongping OuyangZe JiPublished in: J. Intell. Robotic Syst. (2023)
Keyphrases
- reinforcement learning
- state space
- function approximation
- model free
- dynamic programming
- optimal solution
- optimal policy
- deep learning
- machine learning
- data sets
- learning process
- simulated annealing
- temporal difference learning
- reinforcement learning algorithms
- indoor environments
- heuristic function
- greedy heuristic
- tabu search
- search algorithm
- multi agent
- navigation systems
- markov decision process
- beam search
- heuristic solution
- action selection
- heuristic methods
- solution quality
- knapsack problem
- feasible solution
- combinatorial optimization
- markov decision processes
- decision trees