Recursive neural network-based design of unmanned aircraft swarm collaborative mission execution and autonomous navigation system
, , , e
24 mar 2025
INFORMAZIONI SU QUESTO ARTICOLO
Pubblicato online: 24 mar 2025
Ricevuto: 08 nov 2024
Accettato: 19 feb 2025
DOI: https://doi.org/10.2478/amns-2025-0772
Parole chiave
© 2025 Ken Chen et al., published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Layered-RSAC algorithm parameters
| Argument | Value |
|---|---|
| Learning rate | 0.0001 |
| Batch learning scale | 22 |
| Multithread scale | 15 |
| Discount factor | 0.99 |
| Maximum training step size | 186000 |
| Prior Policy experience attenuation | 0.00005 |
| Prior Policy initial variance | 0.15 |
Analysis of UAV autonomous landing accuracy
| Experiment | Landing position/m | Landing deviation/m |
|---|---|---|
| 1 | (7.813,7.930) | (0.187,0.070) |
| 2 | (8.105,8.098) | (0.105,0.098) |
| 3 | (7.983,8.021) | (0.017,0.021) |
| 4 | (7.892,7.928) | (0.108,0.072) |
| 5 | (8.221,8.097) | (0.221,0.097) |
| 6 | (7.779,8.006) | (0.221,0.006) |
| 7 | (8.201,8.195) | (0.201,0.195) |
| 8 | (8.112,8.099) | (0.112,0.099) |
| 9 | (8.250,8.192) | (0.250,0.192) |
| 10 | (8.004,7.995) | (0.004,0.005) |
| 11 | (7.921,8.107) | (0.079,0.107) |
| 12 | (8.112,8.099) | (0.112,0.099) |
| 13 | (7.911,8.024) | (0.089,0.024) |
| 14 | (7.899,8.133) | (0.101,0.133) |
| 15 | (7.980,8.210) | (0.020,0.210) |
| 16 | (7.821,7.905) | (0.179,0.095) |
| 17 | (8.199,8.207) | (0.199,0.207) |
| 18 | (8.002,8.019) | (0.002,0.019) |