A Deep Reinforcement Learning Approach for Path Following on a Quadrotor.
Bartomeu RubíBernardo MorcegoRamon PérezPublished in: ECC (2020)
Keyphrases
- reinforcement learning
- function approximation
- reinforcement learning algorithms
- model free
- semidefinite programming
- state space
- aerial vehicles
- learning algorithm
- markov decision processes
- action selection
- multi agent
- supervised learning
- optimal policy
- obstacle avoidance
- robotic control
- stochastic approximation
- temporal difference
- machine learning
- partially observable
- action space
- optimal control
- transition model
- real time