Eligibility traces in an autonomous soccer robot with obstacle avoidance and navigation policy.
Seyed Omid AzarkasbSeyed Hossein KhastehPublished in: Appl. Soft Comput. (2024)
Keyphrases
- obstacle avoidance
- eligibility traces
- autonomous vehicles
- mobile robot
- path planning
- policy evaluation
- reinforcement learning
- autonomous navigation
- reinforcement learning algorithms
- trajectory planning
- visual navigation
- robotic systems
- unknown environments
- model free
- motion planning
- optimal policy
- robot soccer
- temporal difference
- visually guided
- multi robot
- autonomous robots
- potential field
- markov decision processes
- dynamic environments
- policy iteration
- reinforcement learning methods
- reward function
- least squares
- function approximation
- collision avoidance
- monte carlo
- action selection
- variance reduction
- markov decision process
- neural network
- degrees of freedom
- fuzzy logic
- state space
- artificial neural networks
- learning algorithm