Local 2-D Path Planning of Unmanned Underwater Vehicles in Continuous Action Space Based on the Twin-Delayed Deep Deterministic Policy Gradient.
Zhenzhong ChuYu WangDaqi ZhuPublished in: IEEE Trans. Syst. Man Cybern. Syst. (2024)
Keyphrases
- path planning
- action space
- underwater vehicles
- policy gradient
- single agent
- state action
- reinforcement learning
- path finding
- dynamic environments
- reinforcement learning methods
- state space
- markov decision processes
- function approximators
- mobile robot
- real valued
- reinforcement learning algorithms
- multi robot
- average reward
- optimal path
- function approximation
- multi agent
- motion planning
- action selection
- markov decision process
- stochastic processes
- decision problems
- machine learning
- autonomous vehicles
- mathematical models
- stochastic games
- heuristic search
- unmanned aerial vehicles
- multi agent systems
- finite state
- optimal control
- optimal policy
- learning algorithm