Performance Reliability of Reinforcement Learning Algorithms in Obstacle Avoidance Game with Differing Reward Formulations.
Brandon E. HansenHaley DozierPublished in: CSCI (2022)
Keyphrases
- obstacle avoidance
- reinforcement learning algorithms
- reinforcement learning
- eligibility traces
- reward function
- stochastic games
- mobile robot
- partially observable environments
- path planning
- total reward
- state space
- markov decision processes
- model free
- learning agent
- reinforcement learning problems
- reinforcement learning methods
- dynamic environments
- policy gradient
- average reward
- autonomous vehicles
- reward shaping
- learning algorithm
- function approximation
- motion planning
- optimal policy
- multiagent reinforcement learning
- temporal difference
- agent learns
- multi robot
- nash equilibrium
- fuzzy logic controller
- multiple agents
- fuzzy controller
- inverse reinforcement learning
- sufficient conditions
- particle swarm optimization
- evolutionary algorithm
- neural network