Traffic Flow Optimization on Melbourne Parking Probability 0.2 Real-world data
108.02Avg Time Loss (s)D-PA (Ours)
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| D-PA (Ours)Parking Probability=0.22025.12 | 108.02 | 25.58 | 2.11 | |
| D-PA (DQN)Parking Probability=0.2, Backbone=DQN2025.12 | 121.05 | 16.6 | 1.43 | |
| D-PA (A2C)Parking Probability=0.2, Backbone=A2C2025.12 | 125.34 | 13.64 | 1.5 | |
| D-PA (Du-DQN)Parking Probability=0.2, Backbone=Dueling DQN2025.12 | 129.07 | 11.07 | 1.28 | |
| D-PA (PPO)Parking Probability=0.2, Backbone=PPO2025.12 | 129.49 | 10.78 | 1.29 | |
| C-PAParking Probability=0.22025.12 | 129.82 | 10.56 | 63.52 | |
| D-PA (D-DQN)Parking Probability=0.2, Backbone=Double DQN2025.12 | 131.82 | 9.18 | 1.27 | |
| S-PAParking Probability=0.22025.12 | 133.1 | 8.3 | 1.29 | |
| No-PAParking Probability=0.22025.12 | 145.14 | — | 0.98 |