An Energy-Saving Snake Locomotion Gait Policy Using Deep Reinforcement Learning.
Yilang LiuAmir Barati FarimaniPublished in: CoRR (2021)
Keyphrases
- energy saving
- reinforcement learning
- optimal policy
- legged robots
- energy consumption
- quadruped robot
- policy search
- power consumption
- wireless sensor networks
- markov decision process
- action selection
- robot control
- power saving
- energy conservation
- control policy
- policy gradient
- state space
- actor critic
- partially observable
- energy efficient
- active contours
- global warming
- function approximators
- reward function
- energy management
- save energy
- energy efficiency
- markov decision processes
- action space
- mobile robot
- policy iteration
- agent learns
- reinforcement learning algorithms
- robotic systems
- degrees of freedom
- markov decision problems
- dynamic programming
- multi agent
- gait recognition
- air conditioning
- function approximation
- real robot
- electricity consumption
- temporal difference
- partially observable markov decision processes
- data sets
- rough terrain
- sensor networks
- management system
- optimal control
- data center
- response time