As demand grows for high-altitude, high-speed vehicles operating in complex, contested environments, autonomous navigation under conditions where GNSS satellite signals are unavailable or ground infrastructure is damaged has become a critical technological bottleneck. Existing navigation methods — including visual navigation, terrain matching, inertial navigation, celestial navigation, and geomagnetic navigation — each have inherent limitations, making it difficult to simultaneously satisfy the demands of all-weather operation, high accuracy, and long endurance. A team from the School of Astronautics at Beihang University has proposed a different approach, borrowed from an unexpected source: spacecraft formation flying.
Borrowing a Trick From Spacecraft
In a study published in Space: Science & Technology, the researchers extend the principle of using relative position vectors for absolute navigation — already used in spacecraft formation flying — to high-altitude, high-speed vehicles that are significantly influenced by non-conservative forces such as engine thrust and aerodynamic drag. The method works by using accelerometer measurements from inertial navigation to subtract the effects of those non-conservative accelerations, allowing the gravitational acceleration information that reflects a vehicle’s absolute position to be extracted from the relative position vectors between vehicles flying in formation.
The team built an absolute navigation method that fuses relative position vectors, inertial navigation, and star trackers, formulating the vehicle’s state in the Earth-Centered Earth-Fixed reference frame — comprising position errors, velocity errors, and accelerometer biases — and using relative position vector errors between vehicles as measurements in an extended Kalman filter. Inter-vehicle relative position vectors are obtained through laser ranging and optical direction finding, non-conservative accelerations are measured by accelerometers, and attitude information comes from star trackers; the three data sources are fused to output estimates of position, velocity, and accelerometer biases, without any external signal input.
200 Meters Versus Several Kilometers
Simulation results demonstrate that under a baseline scenario with a formation spacing of 400 kilometers and a flight altitude of 50 kilometers, the method achieves three-dimensional positioning accuracy better than 200 meters over a 150-minute mission. A comparative scheme relying solely on inertial navigation and star trackers, by contrast, diverges to the order of several kilometers within just 10 minutes — confirming the critical role of the relative position vector measurements. Over the final three hours of the baseline simulation, the root-mean-square errors along the three axes were 127.5, 90.4, and 82.6 meters, for a total three-dimensional error of 176.8 meters.
Analysis of the factors influencing performance found that accelerometer measurement noise has the most significant impact: when noise increased from 10 to 30 micrograms per root-hertz, the three-dimensional error grew from 177 to 317 meters. Larger formation spacing improved navigation performance, with errors of 237 meters at 200 kilometers of spacing versus 146 meters at 800 kilometers, while flight altitude between 50 and 200 kilometers had a comparatively minor effect. The researchers say the work offers a feasible technical solution for autonomous navigation of high-altitude, high-speed vehicle formations in GNSS-denied environments, with significant engineering value for enhancing survivability and mission capability in complex, contested scenarios.
The study, “Absolute Navigation of High-Altitude and High-Speed Vehicles Using Relative Position Vector Measurements,” was published in Space: Science & Technology (DOI: 10.34133/space.0402).