CosFly-VLA: Spatially Aware Vision-Language-Action Model for UAV Tracking

Jul 17, 2026

CosFly-VLA is a spatially aware vision-language-action (VLA) model designed for UAV tracking in complex environments with frequent occlusions. The model jointly grounds targets, estimates their visibility, and generates flight actions using a structured prediction interface. Its multi-stage training pipeline includes spatial pretraining, curriculum-based fine-tuning, chain-of-thought reasoning, and reinforcement learning. Compared to OpenVLA, CosFly-VLA achieves over 34% reduction in displacement error and up to 29.8% improvement in success rate on benchmark tests.

Why it matters: This work demonstrates a significant advance in robust UAV tracking by integrating spatial reasoning and occlusion handling, enabling more reliable autonomous navigation in challenging real-world scenarios.

Full story at: arXiv Robotics ↗