Reinforcement Learning for UAV control with Policy and Reward Shaping.
Cristian Millán-AriasRuben ContrerasFrancisco CruzBruno J. T. FernandesPublished in: SCCC (2022)
Keyphrases
- reward shaping
- reinforcement learning
- control policy
- markov decision problems
- control policies
- optimal policy
- action selection
- reinforcement learning algorithms
- policy search
- markov decision process
- complex domains
- state space
- optimal control
- agent learns
- unmanned aerial vehicles
- reward function
- learning algorithm
- control system
- partially observable
- infinite horizon
- partially observable markov decision processes
- policy iteration
- function approximation
- model free
- dynamic programming
- function approximators
- temporal difference
- average reward
- finite state
- continuous state
- policy gradient
- transition model
- markov decision processes