CARL: Controllable Agent with Reinforcement Learning for Quadruped Locomotion.
Ying-Sheng LuoJonathan Hans SoesenoTrista Pei-Chun ChenWei-Chao ChenPublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- multi agent
- action selection
- quadruped robot
- legged robots
- learning agent
- multi agent systems
- multiagent systems
- partially observable
- robot control
- learning capabilities
- intelligent agents
- mobile agents
- multi agent environments
- state action
- reward function
- learning agents
- model free
- exploration strategy
- reinforcement learning agents
- state space
- markov decision process
- single agent
- dynamic environments
- autonomous agents
- machine learning
- decision making
- autonomous learning
- reward shaping
- state abstraction
- degrees of freedom
- agent architecture
- agent learns
- multiple agents
- mobile robot
- decision theoretic
- software agents
- cooperative
- function approximation
- multi agent reinforcement learning
- agent receives
- learning algorithm
- agent systems
- temporal difference
- markov decision processes
- action space
- pedagogical agents
- agent model
- agent technology
- optimal control
- dynamic programming
- control system
- learning process