Research

Aerial autonomy

Uncertainty-Aware Vision-Based Autonomous Aerial Refueling

Minhyuk Jang*, Thinh Nguyen*, Vivek Khatana, Sandeep Banik, Naira Hovakimyan, and Petros G. Voulgaris

Submitted to IEEE ICRA 2027

* Equal contribution

Vision-based docking · FlightGear / JSBSim simulation

Overview

Autonomous probe-and-drogue refueling requires a receiver aircraft to align its probe with a moving drogue. The controller uses margins derived from uncertainty in relative-state estimation and aircraft motion.

The architecture combines vision-based drogue measurements, differential GPS, and avionics in an extended Kalman filter. Conformal calibration characterizes near-contact estimation error, while an offline covariance design assesses position dispersion and corrective-command budgets. Together, they configure an online model predictive controller.

Demonstration

The video shows simulated docking approaches in FlightGear / JSBSim. The study evaluates how geometric and command margins affect crossing accuracy and control effort.