Cross-Entropy Regularized Policy Gradient for Multirobot Nonadversarial Moving Target Search.
Hongliang GuoZhaokai LiuRui ShiWei-Yun YauDaniela RusPublished in: IEEE Trans. Robotics (2023)
Keyphrases
- cross entropy
- multi robot
- moving target search
- policy gradient
- search algorithm
- path planning
- mobile robot
- log likelihood
- maximum likelihood
- reinforcement learning
- optimal control
- function approximation
- gradient method
- robotic systems
- language modeling
- ranking functions
- evaluation metrics
- least squares
- error function
- dynamic environments
- partially observable markov decision processes
- dynamic programming
- simulated annealing