Merging planning in dense traffic scenarios using interactive safe reinforcement learning.
Xiaohui HouMinggang GanWei WuChenyu WangYuan JiShiyue ZhaoPublished in: Knowl. Based Syst. (2024)
Keyphrases
- reinforcement learning
- action selection
- planning problems
- mixed initiative
- deterministic domains
- function approximation
- complex domains
- partially observable
- ai planning
- user friendly
- state space
- user interaction
- macro actions
- stochastic domains
- partial observability
- virtual reality
- model free
- computer graphics
- learning algorithm
- reinforcement learning algorithms
- optimal policy
- markov decision processes
- machine learning
- clustering algorithm
- planning systems
- planning process
- plan generation
- heuristic search
- reinforcement learning problems
- domain independent
- transfer learning