Pursuit-evasion with Decentralized Robotic Swarm in Continuous State Space and Action Space via Deep Reinforcement Learning.
Gurpreet SinghDaniel M. LofaroDonald SofgePublished in: ICAART (1) (2020)
Keyphrases
- pursuit evasion
- action space
- continuous state spaces
- reinforcement learning
- state space
- markov decision processes
- continuous state
- multi agent
- real robot
- cooperative
- real valued
- single agent
- stochastic processes
- function approximators
- control problems
- control policies
- search space
- action selection
- mobile robot
- learning algorithm
- function approximation
- optimal policy
- state variables
- markov decision problems
- partially observable
- robotic systems
- reinforcement learning algorithms
- finite state
- probability distribution
- partially observable markov decision processes
- dynamic environments
- supervised learning
- bayesian networks