Table 2 Algorithm result index comparison in “long corridor”.
From: Improved double DQN with deep reinforcement learning for UAV indoor autonomous obstacle avoidance
Double DQN | Improved Double DQN | Improvement | |
|---|---|---|---|
Average cumulative reward | 17.66 | 21.7 | 22.88% |
Maximum cumulative reward | 51.81 | 104.43 | 101.56% |
Average safe flight distance(m) | 22.7 | 27.96 | 23.17% |
Maximum safe flight distance(m) | 64 | 131.6 | 105.62% |
