DDPG Reinforcement Learning Experiment for Improving the Stability of Bipedal Walking of Humanoid Robots.
Yeonghun ChunJunghun ChoiInjoon MinMinsung AhnJeakweon HanPublished in: SII (2023)
Keyphrases
- reinforcement learning methods
- reinforcement learning
- humanoid robot
- imitation learning
- walking speed
- motion planning
- model free
- biologically inspired
- human robot interaction
- multi modal
- markov decision processes
- state space
- function approximation
- real robot
- machine learning
- motion capture
- temporal difference
- motor control
- feature selection
- stability analysis
- multi agent
- learning algorithm