Multi-UAV roundup strategy method based on deep reinforcement learning CEL-MADDPG algorithm.
Bo LiJianmei WangChao SongZhipeng YangKaifang WanQingfu ZhangPublished in: Expert Syst. Appl. (2024)
Keyphrases
- dynamic programming
- cost function
- preprocessing
- experimental evaluation
- high accuracy
- detection method
- significant improvement
- detection algorithm
- objective function
- computational cost
- high efficiency
- optimization algorithm
- estimation algorithm
- improved algorithm
- segmentation algorithm
- theoretical analysis
- selection algorithm
- convergence rate
- classification method
- k means
- single pass
- computationally efficient
- computational complexity
- clustering method
- matching algorithm
- segmentation method
- similarity measure
- search strategy
- learning algorithm
- synthetic and real images
- support vector machine svm
- reinforcement learning
- input data
- recognition algorithm
- energy function
- computational efficiency
- expectation maximization
- simulated annealing
- probabilistic model
- classification algorithm
- selection strategy
- model free
- optimization method
- em algorithm
- greedy strategy
- policy search
- fitted q iteration
- region of interest
- edge detection
- state space
- face recognition
- feature selection