Neural Networks Versus Conventional Filters for Inertial-Sensor-based Attitude Estimation

Inertial measurement units are commonly used to estimate the attitude of moving objects. Numerous nonlinear filter approaches have been proposed for solving the inherent sensor fusion problem. However, when a large range of different dynamic and static rotational and translational motions is considered, the attainable accuracy is limited by the need for situation-dependent adjustment of accelerometer and gyroscope fusion weights. We investigate to what extent these limitations can be overcome by means of artificial neural networks and how much domain-specific optimization of the neural network model is required to outperform the conventional filter solution. A diverse set of motion recordings with a marker-based optical ground truth is used for performance evaluation and comparison. The proposed neural networks are found to outperform the conventional filter across all motions only if domain-specific optimizations are introduced. We conclude that they are a promising tool for inertial-sensor-based real-time attitude estimation, but both expert knowledge and rich datasets are required to achieve top performance.

[1]  Josef Kittler,et al.  Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[2]  Leslie N. Smith,et al.  Cyclical Learning Rates for Training Neural Networks , 2015, 2017 IEEE Winter Conference on Applications of Computer Vision (WACV).

[3]  Jeremy Howard,et al.  fastai: A Layered API for Deep Learning , 2020, Inf..

[4]  Frank Hutter,et al.  SGDR: Stochastic Gradient Descent with Warm Restarts , 2016, ICLR.

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

[6]  Yi Zheng,et al.  Time Series Classification Using Multi-Channels Deep Convolutional Neural Networks , 2014, WAIM.

[7]  Domenico Formica,et al.  On the Orientation Error of IMU: Investigating Static and Dynamic Accuracy Targeting Human Motion , 2016, PloS one.

[8]  Geoffrey E. Hinton,et al.  Lookahead Optimizer: k steps forward, 1 step back , 2019, NeurIPS.

[9]  Luis Perez,et al.  The Effectiveness of Data Augmentation in Image Classification using Deep Learning , 2017, ArXiv.

[10]  Diganta Misra,et al.  Mish: A Self Regularized Non-Monotonic Neural Activation Function , 2019, ArXiv.

[11]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[12]  Wen Yu,et al.  Non-linear system modeling using LSTM neural networks , 2018 .

[13]  Thomas Seel,et al.  Eliminating the Effect of Magnetic Disturbances on the Inclination Estimates of Inertial Sensors , 2017 .

[14]  Martin Brossard,et al.  RINS-W: Robust Inertial Navigation System on Wheels , 2019, 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[15]  Andrew Markham,et al.  Deep Neural Network Based Inertial Odometry Using Low-Cost Inertial Measurement Units , 2021, IEEE Transactions on Mobile Computing.

[16]  Thomas B. Schön,et al.  Deep Convolutional Networks in System Identification , 2019, 2019 IEEE 58th Conference on Decision and Control (CDC).

[17]  Keyu Wu,et al.  AbolDeepIO: A Novel Deep Inertial Odometry Network for Autonomous Vehicles , 2020, IEEE Transactions on Intelligent Transportation Systems.

[18]  Heiga Zen,et al.  WaveNet: A Generative Model for Raw Audio , 2016, SSW.

[19]  Martin Brossard,et al.  Denoising IMU Gyroscopes With Deep Learning for Open-Loop Attitude Estimation , 2020, IEEE Robotics and Automation Letters.

[20]  Thomas Seel,et al.  Overcoming Bandwidth Limitations in Wireless Sensor Networks by Exploitation of Cyclic Signal Patterns: An Event-triggered Learning Approach , 2020, Sensors.

[21]  Diganta Misra.  Mish: A Self Regularized Non-Monotonic Activation Function , 2020, BMVC.

[22]  Xiaodong Cui,et al.  Data Augmentation for Deep Neural Network Acoustic Modeling , 2015, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[23]  Richard Socher,et al.  Regularizing and Optimizing LSTM Language Models , 2017, ICLR.

[24]  Bertram Taetz,et al.  IMU-to-Segment Assignment and Orientation Alignment for the Lower Body Using Deep Learning , 2018, Sensors.

[25]  Alain Pagani,et al.  Learning to Fuse: A Deep Learning Approach to Visual-Inertial Camera Pose Estimation , 2016, 2016 IEEE International Symposium on Mixed and Augmented Reality (ISMAR).

[26]  Kai-Wei Chiang,et al.  An Artificial Neural Network Embedded Position and Orientation Determination Algorithm for Low Cost MEMS INS/GPS Integrated Sensors , 2009, Sensors.

[27]  A. Cereatti,et al.  Accuracy of the Orientation Estimate Obtained Using Four Sensor Fusion Filters Applied to Recordings of Magneto-Inertial Sensors Moving at Three Rotation Rates , 2019, 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[28]  Liyuan Liu,et al.  On the Variance of the Adaptive Learning Rate and Beyond , 2019, ICLR.

[29]  Yann Ollivier,et al.  Unbiasing Truncated Backpropagation Through Time , 2017, ArXiv.

[30]  Max Jaderberg,et al.  Population Based Training of Neural Networks , 2017, ArXiv.