AeroRSI: Recursive Self-Improvement for Aerial Vision-Language Navigation
Organizations: SIGS, Tsinghua University · Department of Electronic Engineering, Tsinghua University · Foshan University · South China University of Technology
Abstract
Vision-language-action (VLA) models enable unmanned aerial vehicles (UAVs) to follow natural-language instructions, yet still fail frequently in closed-loop flight. Correcting these failures requires repairs, i.e., actions that avoid them at the states the policy itself reaches, which human pilots cannot label at scale and rule-based planners cannot supply reliably. In this paper, we present AeroRSI, a recursive self-improvement framework in which an aerial navigation policy iteratively learns from its own failures. Our key insight is that a failing policy should judge its own repairs by flying on from them. Judged in this way, its sampled actions and even rule-based proposals that hurt it when imitated directly become reliable supervision. Moreover, an episode terminates at its first failure, concealing subsequent failures until the earlier ones are repaired. AeroRSI thus renews its supervision in every round, anchoring the failures of the current policy, verifying repairs by its continuation, and distilling them while keeping its behavior elsewhere. On TravelUAV, three rounds raise the success rate of a supervised 2B-parameter policy from 46.4% to 67.5% and reduce collisions by half, surpassing the previous state of the art by 10.6 points from the same initialization without extra human labels or inference cost. Code and models are available at https://github.com/starsholic/AeroRSI.
Figures & tables
| Policy | Initialized from | Trained on |
|---|---|---|
| Supervised policy (SFT) [ 1 ] | Qwen3-VL-2B [ 104 ] | Expert demonstrations |
| AeroDPO [ 1 ] | SFT | Collision corrections |
| AeroRSI (SFT init.) | SFT | Verified repairs |
| AeroRSI (DPO init.) | AeroDPO | Verified repairs |
| Method | Guide | Seen | Unseen Object | Unseen Map | Overall | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| NE | SR | OSR | SPL | NE | SR | OSR | SPL | NE | SR | OSR | SPL | SR | ||
| TravelUAV [ 3 ] | L1 | 106.28 | 16.10 | 44.26 | 14.30 | 118.11 | 22.42 | 46.90 | 20.51 | 138.80 | 4.18 | 20.77 | 3.84 | 13.62 |
| NavFoM [ 8 ] | L1 | 93.05 | 29.17 | 49.24 | 25.03 | 108.04 | 29.83 | 47.99 | 27.20 | 125.10 | 6.30 | 18.95 | 5.68 | 22.02 |
| LongFly [ 17 ] | L1 | 60.02 | 36.39 | 65.87 | 31.07 | 66.74 | 43.87 | 64.56 | 38.39 | 108.32 | 11.27 | 30.27 | 9.32 | 29.95 |
| NeuroKalman [ 58 ] | L1 | 71.56 | 25.86 | 58.73 | 22.43 | 71.01 | 32.48 | 60.82 | 28.50 | 100.32 | 8.34 | 34.15 | 7.12 | 21.66 |
| AerialVLA [ 9 ] | FD | 65.88 | 47.96 | 57.69 | 38.54 | 61.45 | 56.60 | 64.86 | 46.61 | 67.42 | 37.58 | 52.92 | 28.22 | 46.46 |
| Round | Seen | UO | UM | SR | SPL | CR | Train SR |
|---|---|---|---|---|---|---|---|
| AeroRSI (SFT init.) | |||||||
| 0 | 48.38 | 56.44 | 36.85 | 46.39 | 37.37 | 38.64 | 45.24 |
| 1 | 62.62 | 67.89 | 43.42 | 57.60 | 46.96 | 29.48 | 61.41 |
| 2 | 73.91 | 79.01 | 45.82 | 66.02 | 51.96 | 21.26 | 71.41 |
| 3 | 75.32 | 82.19 | 46.24 | 67.49 | 51.00 | 19.10 | 70.85 |
| AeroRSI (DPO init.) | |||||||
Appendix figures & tables9 assets
Supplementary material from the paper’s appendix.
Appendix
| Component | Setting |
|---|---|
| Policy | Qwen3-VL-2B; mosaic of the front and downward views; forward, vertical, and yaw components with 99 bins each, plus landing |
| Training | LoRA of rank 64 on all linear layers of the language model, and the visual merger; 2 epochs per round; learning rate ; batch size 16 |
| AFA | Offsets of 1, 2, 4, 8, and 16 actions; 2,400 anchors per round; horizon of actions for offset , at most 20 |
| CVR | samples at temperature 1.0 and three rule-based proposals; , ; minimum progress of 4 m; margin m; two confirmations at most |
| RRL | on 1,990 retention states; replay share from round 2 |
| Round | 1 | 2 | 3 |
|---|---|---|---|
| Flights of the training routes | |||
| Flights | 7,611 | 7,611 | 7,611 |
| Actions (thousands) | 345 | 376 | 435 |
| Verification | |||
| Flights of candidates | 15,237 | 13,975 | 12,451 |
| Confirmation flights | 1,484 | 1,342 | 1,165 |
| Round | Collision | Wrong way | Stuck | Early landing | Timeout |
|---|---|---|---|---|---|
| AeroRSI (SFT init.) | |||||
| 0 | 38.6 | 7.8 | 4.2 | 1.8 | 1.2 |
| 1 | 29.5 | 3.8 | 4.2 | 2.2 | 2.8 |
| 2 | 21.3 | 2.6 | 2.8 | 2.3 | 5.0 |
| 3 | 19.1 | 2.1 | 2.7 | 1.9 | 6.8 |
| AeroRSI (DPO init.) | |||||
| Comparison | SR | |
| Rounds and baselines | ||
| Round 1 vs. round 0 | +11.21 | 10.4 |
| Round 2 vs. round 1 | +8.42 | 8.1 |
| Round 3 vs. round 2 | +1.46 | 1.6 |
| Round 3 vs. round 0 | +21.10 | 18.5 |
| Round 3 vs. AeroDPO | +10.62 | 10.2 |
| Round | 1 | 2 | 3 |
|---|---|---|---|
| Anchors | 2,390 | 2,390 | 2,399 |
| Share of all anchors (%) | |||
| Repair accepted | 43.8 | 38.3 | 34.3 |
| Repair accepted, DPO init. | 46.1 | 38.8 | 33.9 |
| Share of the anchors whose failure reproduces (%) | |||
| Best of four samples qualifies | 66.2 | 68.3 | 68.3 |
| Round | Sampled actions | Potential field | Translation | Climb |
|---|---|---|---|---|
| 1 | 39.2 | 60.7 | 59.3 | 47.6 |
| 2 | 44.3 | 55.4 | 59.6 | 46.2 |
| 3 | 51.0 | 63.6 | 59.0 | 51.2 |
| Anchors repaired by | @1 | @4 | @1 | @4 | @1 | @4 |
|---|---|---|---|---|---|---|
| All (100%) | 33.4 | 61.7 | 63.2 | 83.6 | 73.8 | 86.6 |
| A sampled action (66%) | 45.6 | 79.9 | 71.7 | 90.9 | 80.4 | 92.2 |
| Only a proposal (21%) | 11.0 | 30.4 | 62.1 | 84.3 | 76.7 | 87.9 |
| Neither (13%) | 6.5 | 18.9 | 20.9 | 44.3 | 34.6 | 55.1 |
| Held-out route groups | ||||||
| Continued by | again | |||
|---|---|---|---|---|
| Verified repair | 81.5 | 86.9 | 87.0 | 89.1 |
| held out (211) | 82.5 | 88.2 | 91.9 | 92.9 |
| from sampled actions (460) | 77.6 | 83.3 | 85.4 | 88.0 |
| from proposals (588) | 84.5 | 89.8 | 88.2 | 90.0 |
| Failed action | 26.5 | 57.0 | 63.3 | 66.6 |
| Flights | Routes | Within 10 m | Over 50 m | Median |
|---|---|---|---|---|
| , then | 2,125 | 27.9 | 44.8 | 40.2 m |
| , then | 1,333 | 33.0 | 37.6 | 28.0 m |
| One policy twice | 56 | 50.0 | 30.4 | 12.7 m |
Aerial VLN
Aerial vision-and-language navigation (VLN) enables UAVs to follow natural-language goals using visual perception and flight control.
See all 71 papers