Automated Driving Highway Traffic Merging using Deep Multi-Agent Reinforcement Learning in Continuous State-Action Spaces.
Larry SchesterLuis E. OrtizPublished in: IV (2021)
Keyphrases
- continuous state
- action space
- reinforcement learning
- state space
- robot navigation
- markov decision processes
- real valued
- finite state
- continuous action
- control policies
- action selection
- planning problems
- state dependent
- policy search
- partially observable markov decision processes
- state action
- stochastic processes
- stochastic games
- function approximation
- markov chain
- dynamic programming