Deep learning-based inertia tensor identification of the combined spacecraft

Abstract

The identification accuracy of inertia tensor of combined spacecraft, which is composed by a servicing spacecraft and a captured target, could be easily affected by the measurement noise of angular rate. Due to frequently changing operating environments of combined spacecraft in space, the measurement noise of angular rate can be very complex. In this paper, an inertia tensor identification approach based on deep learning method is proposed to improve the ability of identifying inertia tensor of combined spacecraft in the presence of complex measurement noise. A deep neural network model for identification is constructed and trained by enough training data and a designed learning strategy. To verify the identification performance of the proposed deep neural network model, two testing set with different ranks of measure noises are used for simulation tests. Comparison tests are also delivered among the proposed deep neural network model, recursive least squares identification method, and tradition deep neural network model. The comparison results show that the proposed deep neural network model yields a more accurate and stable identification performance for inertia tensor of combined spacecraft in changeable and complex operating environments.

Deep Learning and Artificial Neural Networks for Spacecraft Dynamics, Navigation and Control

“…The angular rates and control torques of combined spacecraft are set as the input of a deep neural network model, and conversely, the inertia tensor is then set as the output. Training the MLP model refers to the process of extracting a higher abstract feature, i.e., the inertia tensor [36,37]. Another approach to solve the parametric reconstruction is to use a recurrent neural network.…”

Autonomous trajectory planning method for hypersonic vehicles in glide phase based on DDPG algorithm

“…12 As the internal calculation of DNN involves simple matrix multiplications, control values are output in real time by online sensing of the current flight state, which is then input into the DNN. 13 Chai et al 14 reported an optimal trajectory planning and real-time altitude control method for a six degrees of freedom spacecraft based on DNN and proved the real-time ability and reliability of the method. Yang Shi et al 15 used a homotopy algorithm to generate a large number of state-action pairs and trained a DNN offline for online trajectory generation.…”