Static Attitude Determination Using Convolutional Neural Networks

The need to estimate the orientation between frames of reference is crucial in spacecraft navigation. Robust algorithms for this type of problem have been built by following algebraic approaches, but data-driven solutions are becoming more appealing due to their stochastic nature. Hence, an approach based on convolutional neural networks in order to deal with measurement uncertainty in static attitude determination problems is proposed in this paper. PointNet models were trained with different datasets containing different numbers of observation vectors that were used to build attitude profile matrices, which were the inputs of the system. The uncertainty of measurements in the test scenarios was taken into consideration when choosing the best model. The proposed model, which used convolutional neural networks, proved to be less sensitive to higher noise than traditional algorithms, such as singular value decomposition (SVD), the q-method, the quaternion estimator (QUEST), and the second estimator of the optimal quaternion (ESOQ2).

[1]  Adrian Munteanu,et al.  Learning to Estimate the Body Shape Under Clothing From a Single 3-D Scan , 2021, IEEE Transactions on Industrial Informatics.

[2]  Ledan Qian,et al.  Sequence-Dropout Block for Reducing Overfitting Problem in Image Classification , 2020, IEEE Access.

[3]  Han Wang,et al.  OriNet: Robust 3-D Orientation Estimation With a Single Particular IMU , 2020, IEEE Robotics and Automation Letters.

[4]  Yi Zhou,et al.  On the Continuity of Rotation Representations in Neural Networks , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Chyi-Yeu Lin,et al.  6D pose estimation with combined deep learning and 3D vision techniques for a fast and accurate object grasping , 2021, Robotics Auton. Syst..

[6]  Yi-Qing Ni,et al.  Improved Most Likely Heteroscedastic Gaussian Process Regression via Bayesian Residual Moment Estimator , 2020, IEEE Transactions on Signal Processing.

[7]  Jin Wu,et al.  Rigid 3-D Registration: A Simple Method Free of SVD and Eigendecomposition , 2020, IEEE Transactions on Instrumentation and Measurement.

[8]  Aleksandr Cariow,et al.  An Algorithm for Quaternion–Based 3D Rotation , 2020 .

[9]  P. B. Davenport.  A vector approach to the algebra of rotations with applications , 1968 .

[10]  Feng Li,et al.  Intelligent active fault-tolerant system for multi-source integrated navigation system based on deep neural network , 2019, Neural Computing and Applications.

[11]  Jian Xie,et al.  Accurate Direct Strapdown Direction Cosine Algorithm , 2019, IEEE Transactions on Aerospace and Electronic Systems.

[12]  M. Shuster,et al.  Three-axis attitude determination from vector observations , 1981 .

[13]  Friedrich Fraundorfer,et al.  Efficient Ego-Motion Estimation for Multi-Camera Systems With Decoupled Rotation and Translation , 2020, IEEE Access.

[14]  M. Franaszek,et al.  Orientation Uncertainty Characteristics of Some Pose Measuring Systems , 2017, Mathematical problems in engineering.

[15]  M. D. Shuster.  Approximate algorithms for fast optimal attitude computation , 1978 .

[16]  Yongbin Yu,et al.  RMAF: Relu-Memristor-Like Activation Function for Deep Learning , 2020, IEEE Access.

[17]  J. Keat.  Analysis of Least-Squares Attitude Determination Routine DOAO , 1977 .

[18]  Ademir Nied,et al.  Hybrid deep learning for power generation forecasting in active solar trackers , 2020, IET Generation, Transmission & Distribution.

[19]  Mingqiang Wei,et al.  Sparse-to-Dense Multi-Encoder Shape Completion of Unstructured Point Cloud , 2020, IEEE Access.

[20]  F. Markley.  Attitude determination using vector observations and the singular value decomposition , 1988 .

[21]  H. Vincent Poor,et al.  On Estimating the Norm of a Gaussian Vector Under Additive White Gaussian Noise , 2019, IEEE Signal Processing Letters.

[22]  G. Wahba.  A Least Squares Estimate of Satellite Attitude , 1965 .

[23]  Jong-Hwan Kim,et al.  D3PointNet: Dual-Level Defect Detection PointNet for Solder Paste Printer in Surface Mount Technology , 2020, IEEE Access.

[24]  Fu-Kwun Wang,et al.  Long Short-Term Memory With Attention Mechanism for State of Charge Estimation of Lithium-Ion Batteries , 2020, IEEE Access.

[25]  Ali Khenchaf,et al.  Robust-Extended Kalman Filter and Long Short-Term Memory Combination to Enhance the Quality of Single Point Positioning , 2020 .

[26]  Ian D. Reid,et al.  Deep-6DPose: Recovering 6D Object Pose from a Single RGB Image , 2018, ArXiv.

[27]  Gang Wang,et al.  A Method of Robot Base Frame Calibration by Using Dual Quaternion Algebra , 2018, IEEE Access.

[28]  Daniele Mortari,et al.  Second Estimator of the Optimal Quaternion , 2000 .

[29]  F. Markley.  Attitude determination using vector observations: A fast optimal matrix algorithm , 1993 .

[30]  Karel Zikan,et al.  A note on averaging rotations , 1993, Proceedings of IEEE Virtual Reality Annual International Symposium.

[31]  Jianwei Zhang,et al.  Occlusion Resistant Object Rotation Regression from Point Cloud Segments , 2018, ECCV Workshops.

[32]  Dieter Fox,et al.  PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes , 2017, Robotics: Science and Systems.

[33]  Jiayong Ye,et al.  Graph neural network for 6D object pose estimation , 2021, Knowl. Based Syst..

[34]  Daniele Mortari,et al.  A Survey on Star Identification Algorithms , 2009, Algorithms.

[35]  Juan Hu,et al.  Using Deep Learning in Semantic Classification for Point Cloud Data , 2019, IEEE Access.

[36]  Guillermo Heredia,et al.  Design of an Anthropomorphic, Compliant, and Lightweight Dual Arm for Aerial Manipulation , 2018, IEEE Access.

[37]  Wei Wu,et al.  PointCNN: Convolution On X-Transformed Points , 2018, NeurIPS.

[38]  Shang Gao,et al.  Separating the Structural Components of Maize for Field Phenotyping Using Terrestrial LiDAR Data and Deep Convolutional Neural Networks , 2020, IEEE Transactions on Geoscience and Remote Sensing.