Reinforcement Learning-Based Autonomous Navigation and Obstacle Avoidance for USVs under Partially Observable Conditions

Figure 3

The MDP framework for USV to interact with environments.