Deep reinforcement learning for tensegrity robot locomotion.
Marvin ZhangXinyang GengJonathan BruceKen CaluwaertsMassimo VespignaniVytas SunSpiralPieter AbbeelSergey LevinePublished in: ICRA (2017)
Keyphrases
- robot control
- reinforcement learning
- mobile robot
- rough terrain
- robotic systems
- real robot
- legged robots
- central pattern generator
- autonomous robots
- robot moves
- autonomous learning
- path planning
- function approximation
- legged locomotion
- humanoid robot
- obstacle avoidance
- perceptual aliasing
- state space
- visual servoing
- human robot interaction
- goal directed
- mobile robotics
- autonomous navigation
- robot manipulators
- robot behavior
- degrees of freedom
- unknown environments
- human robot
- quadruped robot
- machine learning
- deep learning
- temporal difference
- motion planning
- multi robot
- transfer learning
- dynamic programming
- reinforcement learning algorithms
- search and rescue
- robot navigation
- experimental platform
- markov decision processes
- multi modal
- neural network