How to train your robot with deep reinforcement learning: lessons we have learned.
Julian IbarzJie TanChelsea FinnMrinal KalakrishnanPeter PastorSergey LevinePublished in: Int. J. Robotics Res. (2021)
Keyphrases
- reinforcement learning
- robot control
- real robot
- mobile robot
- perceptual aliasing
- human robot interaction
- function approximation
- autonomous learning
- vision system
- robot navigation
- path planning
- lessons learned
- imitation learning
- state space
- robotic systems
- humanoid robot
- autonomous robots
- sensory inputs
- simulated robot
- learned knowledge
- markov decision processes
- multi robot
- obstacle avoidance
- learning algorithm
- human teacher
- optimal policy
- robot behavior
- dynamic programming
- model free
- human robot
- training stage
- service robots
- robot manipulators
- reinforcement learning algorithms
- action selection
- goal directed
- multiple robots
- control signals
- transfer learning
- learning process
- primitive actions