An Index Policy Based on Sarsa and Q-learning for Heterogeneous Smart Target Tracking.
Yuhang HaoZengfu WangJing FuQuan PanPublished in: CoRR (2024)
Keyphrases
- target tracking
- action selection
- reinforcement learning
- function approximation
- optimal policy
- policy iteration
- reinforcement learning algorithms
- function approximators
- kalman filter
- data fusion
- markov decision processes
- rl algorithms
- model free
- reward function
- temporal difference learning
- multi sensor
- video camera
- markov decision process
- mean shift
- state space
- temporal difference
- infrared imagery
- multiple target tracking
- learning algorithm
- cluttered environments
- multi agent
- single agent
- data association
- moving target
- particle filter
- control policy
- average reward
- object tracking
- reinforcement learning methods
- dynamic programming
- learning agent
- image fusion
- low cost
- image sequences