Energy-Optimal Flight Strategy for Solar-Powered Aircraft Using Reinforcement Learning With Discrete Actions.
Wenjun NiDi WuXiaoping MaPublished in: IEEE Access (2021)
Keyphrases
- reinforcement learning
- flight data
- optimal strategy
- optimal control
- action sets
- civil aviation
- dynamic programming
- solar energy
- action selection
- partially observable domains
- initially unknown
- air traffic control
- state space
- continuous state spaces
- function approximation
- situation calculus
- optimal solution
- crew scheduling
- perceptual aliasing
- control policy
- continuous state
- minimum energy
- reward function
- markov decision processes
- learning algorithm
- partial observability
- sensory inputs
- state action
- machine learning
- action space
- temporal difference
- energy minimization
- multi agent
- flight path