TY - GEN
T1 - Online Intelligent Identification Method of Spacecraft Mass Characteristics
AU - Cai, Zhenda
AU - Xu, Xingguo
AU - Guo, Jiahua
AU - Shao, Zhijie
AU - Ma, Guangcheng
AU - Xia, Hongwei
N1 - Publisher Copyright:
© 2023 Technical Committee on Control Theory, Chinese Association of Automation.
PY - 2023
Y1 - 2023
N2 - This paper investigates an online intelligent identification method for spacecraft mass characteristics. Firstly, the basic recursive least square method is given based on the modeling of the air-bearing tested system. To improve the identification speed and accuracy, an intelligent identification method using deep neural networks (DNNs) is introduced, where the training data for DNNs is derived from the designed identification scheme based on the grey wolf algorithm. Then, a reasonable stack mechanism with a small-sample data update is proposed to meet the needs of online identification. Simulation results show the online identification method can achieve high-precision and fast identification with a rotational inertia error of less than 0.33% and a center of mass offset error of less than 1.9%.
AB - This paper investigates an online intelligent identification method for spacecraft mass characteristics. Firstly, the basic recursive least square method is given based on the modeling of the air-bearing tested system. To improve the identification speed and accuracy, an intelligent identification method using deep neural networks (DNNs) is introduced, where the training data for DNNs is derived from the designed identification scheme based on the grey wolf algorithm. Then, a reasonable stack mechanism with a small-sample data update is proposed to meet the needs of online identification. Simulation results show the online identification method can achieve high-precision and fast identification with a rotational inertia error of less than 0.33% and a center of mass offset error of less than 1.9%.
KW - Air bearing testbed
KW - Data stack
KW - Deep neural networks
KW - Grey wolf algorithm
KW - Mass characteristics
KW - Online identification
UR - https://www.scopus.com/pages/publications/85175569254
U2 - 10.23919/CCC58697.2023.10241213
DO - 10.23919/CCC58697.2023.10241213
M3 - 会议稿件
AN - SCOPUS:85175569254
T3 - Chinese Control Conference, CCC
SP - 1391
EP - 1396
BT - 2023 42nd Chinese Control Conference, CCC 2023
PB - IEEE Computer Society
T2 - 42nd Chinese Control Conference, CCC 2023
Y2 - 24 July 2023 through 26 July 2023
ER -