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

Table 2

The obstacle avoidance success rate of the UANOA algorithm and comparison algorithms in three different ocean scenarios.

Environment type	Primary scenario	Medium scenario	Advanced scenario
UANOA success rate	95%	86%	67%
DQN success rate	91%	79%	49%
Random policy success rate	0	0	0