Mass-center location identification of large space debris based on binocular vision and inertial measurement units fusion

Yao Jinming,Li Guangping,Zhang Huibo,Tian Hao,You Bindi,Dai Shijie
DOI: https://doi.org/10.16708/j.cnki.1000-758X.2022.0013
2022-01-01
Chinese space science and technology
Abstract:The mass center of the large free-tumbling debris is the relatively fixed point in the on-orbit operation frame, and is also the benchmark for the conversion of dynamic parameters from the debris conjoined base to the satellite coordinate system. The precise identification of mass-center location is the key to improving the identification accuracy of debris dynamic parameters. A method for the mass-center-location identification of large space debris was proposed based on inertial measurement units and binocular vision. Based on the torque-free Euler equation and dyadic coordinate transformation, the conversion relationship between the inertial measurement units attached to the surface of the space debris was established. Based on the conversion relationship of the inertial measurement units, the redundant inertial measurement units' data was denoised. Using the denoising data, the distances from the inertial measurement units to the mass center were optimized. Using the binocular vision, the dynamic coordinates of the markers on the inertial unit were obtained. Then the distances from the inertial measurement units to the mass center were used, and the large space debris' mass-center location was identified. Simulations with Gaussian white noise data of inertial measurement unit and binocular vision were carried out, and the results show that the error of the inertial measurement units' data is reduced to less than 1% after real-time denoising and that the mass-center-location triaxial error is less than 0. 47 mm. Ground experiments were carried out, and the results show that the triaxial error of the mass-center-location is less than 0. 49 mm. Simulations and experiments prove that the method can provide a more accurate data benchmark for the de-tumbling and capture mission of large space debris.
What problem does this paper attempt to address?