Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2604.19059 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913050860191744 |
|---|---|
| author | Lyu, Lingxue |
| author_facet | Lyu, Lingxue |
| contents | Language-guided unmanned aerial vehicles (UAVs) often fail not from bad reasoning or perception, but from execution
mismatch: the gap between a planned trajectory and the controller's ability to track it when the real dynamics differ
from training (mass changes, drag shifts, actuator delay, wind). We propose AeroBridge-TTA, a language-conditioned
control pipeline that targets this gap with test-time adaptation. It has three parts: a language encoder that maps the
command into a subgoal, an adaptive policy conditioned on the subgoal and a learned latent, and a test-time
adaptation (TTA) module that updates the latent online from observed transitions. On five language-conditioned UAV
tasks under 13 mismatch conditions with the same domain randomization, AeroBridge-TTA ties a strong PPO-MLP baseline
in-distribution and wins all 5 out-of-distribution (OOD) conditions, +22.0 pts on average (62.7% vs. 40.7%); the +8.5
pt overall gain comes entirely from the OOD regime. A same-weights ablation that only changes the step size $α$
shows the latent update itself is responsible for a $4.6\times$ OOD lift. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19059 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AeroBridge-TTA: Test-Time Adaptive Language-Conditioned Control for UAVs Lyu, Lingxue Robotics Language-guided unmanned aerial vehicles (UAVs) often fail not from bad reasoning or perception, but from execution mismatch: the gap between a planned trajectory and the controller's ability to track it when the real dynamics differ from training (mass changes, drag shifts, actuator delay, wind). We propose AeroBridge-TTA, a language-conditioned control pipeline that targets this gap with test-time adaptation. It has three parts: a language encoder that maps the command into a subgoal, an adaptive policy conditioned on the subgoal and a learned latent, and a test-time adaptation (TTA) module that updates the latent online from observed transitions. On five language-conditioned UAV tasks under 13 mismatch conditions with the same domain randomization, AeroBridge-TTA ties a strong PPO-MLP baseline in-distribution and wins all 5 out-of-distribution (OOD) conditions, +22.0 pts on average (62.7% vs. 40.7%); the +8.5 pt overall gain comes entirely from the OOD regime. A same-weights ablation that only changes the step size $α$ shows the latent update itself is responsible for a $4.6\times$ OOD lift. |
| title | AeroBridge-TTA: Test-Time Adaptive Language-Conditioned Control for UAVs |
| topic | Robotics |
| url | https://arxiv.org/abs/2604.19059 |