Reinforcement Learning Controllers for Soft Robots Using Learned Environments.
Uljad BerdicaMatthew JacksonNiccolò Enrico VeroneseJakob N. FoersterPerla MaiolinoPublished in: RoboSoft (2024)
Keyphrases
- reinforcement learning
- robot behavior
- robotic systems
- unstructured environments
- real robot
- robot control
- autonomous robots
- robotic control
- evolutionary robotics
- imitation learning
- multi robot
- state space
- mobile robot
- learned knowledge
- function approximation
- markov decision processes
- human operators
- previously learned
- cooperative
- physical objects
- action selection
- dynamic environments
- real world environments
- learning algorithm
- multi agent environments
- human teacher
- service robots
- robot teams
- reinforcement learning algorithms
- dynamic programming
- learning process
- real world
- legged robots
- model free
- policy search
- complex environments
- optimal control
- optimal policy
- robotic agents
- artificial agents
- control policies
- reward function
- human robot interaction
- neural network