Generalize Robot Learning From Demonstration to Variant Scenarios With Evolutionary Policy Gradient.
Junjie CaoWeiwei LiuYong LiuJian YangPublished in: Frontiers Neurorobotics (2020)
Keyphrases
- policy gradient
- mobile robot
- actor critic
- parametric optimization
- genetic algorithm
- path planning
- optimal control
- evolutionary computation
- variance reduction
- gradient method
- reinforcement learning
- model free reinforcement learning
- function approximation
- average reward
- robot arm
- approximation methods
- reinforcement learning algorithms
- real robot
- radial basis function