Improving Accuracy in Inertial Navigation Systems with Machine Learning

Year 2023, Volume: 15 Issue: 1, 286 - 296, 31.01.2023

Fatih Şahin , Faruk Ulamış

https://doi.org/10.29137/umagd.1129097

Cited By: 1

Abstract

Inertial navigation systems help various air, land and sea vehicles to find their positions by using the sensor data you receive from a unit that is usually configured as an Inertial Measurement Unit (ICU). Recently, this technology has become wearable by integrating into the feet or various parts of the body, but the biggest disadvantage of these systems is that they create errors that increase over time due to the sensors used. Minimizing these errors is of great importance in terms of location accuracy. In inertial navigation systems (ANS) made with an inertial measurement unit mounted on the foot, the correct determination of the zero velocity detection (SHA) process is the most important factor reducing the measurement errors. In this study, long short-term memory (LSTM), a Recurrent Neural Network (RNN) method, was used to detect SHA more accurately. This method makes a binary classification for zero velocity detection using sensor data. ANS measurements made with the proposed method have been applied for different environments and it has been observed that it makes measurements with higher precision than standard ANS.

Keywords

Inertial Navigation, Inertial Sensing, Inertial Measurement Unit, Dead Calculation, Zero Velocity Detection, Machine Learning

Makine Öğrenmesi İle Ataletsel Navigasyon Sistemlerinde Doğruluğun Geliştirilmesi

Abstract

Ataletsel navigasyon sistemleri, genellikle Ataletsel Ölçüm Birimi(AÖB) olarak yapılandırılan bir birimden aldığı sensör verilerini kullanarak hava, kara ve deniz araçlarının konumlarını bulabilmesine yardımcı olmaktadır. Son dönemlerde bu teknoloji ayağa veya vücudun çeşitleri yerlerine entegre edilerek giyilebilir hale getirilmektedir, fakat bu sistemlerin en büyük dezavantajı kullanılan sensörler nedeniyle zamanla artan hatalar oluşturmalarıdır. Bu hataları minimize etmek konum doğruluğu açısından büyük önem taşımaktadır. Ayağa takılı ataletsel ölçüm birimi ile yapılan ataletsel navigasyon sistemlerinde (ANS), sıfır hız algılama (SHA) işleminin doğru tespit edilmesi ölçüm hatalarını düşüren en önemli etkendir. Bu çalışmada, SHA'yı daha doğru bir şekilde tespit etmek için Tekrarlayan Sinir Ağı (RNN) yöntemi olan uzun kısa süreli bellek (LSTM) kullanılmıştır. Bu yöntem sensör verilerini kullanarak sıfır hız algılama için ikili bir sınıflandırma yapmaktadır. Önerilen yöntemle yapılan ANS ölçümleri farklı ortamlar için uygulanmış ve standart ANS' den daha yüksek hassasiyette ölçümler yaptığı görülmüştür.

Keywords

Ataletsel Algılama, Ataletsel Ölçüm Birimi, Ölü Hesaplama, Sıfır Hız Algılama, Makine Öğrenimi, Ataletsel Navigasyon

References

• Curran, K., & Norrby, S. (2009). RFID-enabled location determination within indoor environments. International Journal of Ambient Computing and Intelligence (IJACI), 1(4), 63-86.
• Gu, Y., Lo, A., & Niemegeers, I. (2009). A survey of indoor positioning systems for wireless personal networks. IEEE Communications surveys & tutorials, 11(1), 13-32.
• Fu, Q., & Retscher, G. (2009). Another look indoors GPS+ RFID. GPS World, 20(3).
• Fischer, C., & Gellersen, H. (2010). Location and navigation support for emergency responders: A survey. IEEE Pervasive Computing, 9(01), 38-47.
• Skog, I., Handel, P., Nilsson, J. O., & Rantakokko, J. (2010). Zero-velocity detection—An algorithm evaluation. IEEE transactions on biomedical engineering, 57(11), 2657-2666.
• Foxlin, E. (2005). Pedestrian tracking with shoe-mounted inertial sensors. IEEE Computer graphics and applications, 25(6), 38-46.
• Nilsson, J. O., Skog, I., & Händel, P. (2012). A note on the limitations of ZUPTs and the implications on sensor error modeling. In 2012 International Conference on Indoor Positioning and Indoor Navigation (IPIN), 13-15th November 2012.
• Wahlström, J., Skog, I., Gustafsson, F., Markham, A., & Trigoni, N. (2019). Zero-velocity detection—A Bayesian approach to adaptive thresholding. IEEE Sensors Letters, 3(6), 1-4.
• Tian, X., Chen, J., Han, Y., Shang, J., & Li, N. (2016). A novel zero velocity interval detection algorithm for self-contained pedestrian navigation system with inertial sensors. Sensors, 16(10), 1578.
• Walder, U., & Bernoulli, T. (2010, September). Context-adaptive algorithms to improve indoor positioning with inertial sensors. In 2010 International Conference on Indoor Positioning and Indoor Navigation (pp. 1-6). IEEE.
• Ren, M., Pan, K., Liu, Y., Guo, H., Zhang, X., & Wang, P. (2016). A novel pedestrian navigation algorithm for a foot-mounted inertial-sensor-based system. Sensors, 16(1), 139.
• Ma, M., Song, Q., Li, Y., & Zhou, Z. (2017, December). A zero velocity intervals detection algorithm based on sensor fusion for indoor pedestrian navigation. In 2017 IEEE 2nd Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) (pp. 418-423). IEEE.
• Nilsson, J. O., Gupta, A. K., & Händel, P. (2014, October). Foot-mounted inertial navigation made easy. In 2014 International Conference on Indoor Positioning and Indoor Navigation (IPIN) (pp. 24-29). IEEE.
• Skog, I., Nilsson, J. O., & Händel, P. (2010, September). Evaluation of zero-velocity detectors for foot-mounted inertial navigation systems. In 2010 International Conference on indoor positioning and indoor navigation (pp. 1-6). IEEE.
• Olivares, A., Ramírez, J., Górriz, J. M., Olivares, G., & Damas, M. (2012). Detection of (in) activity periods in human body motion using inertial sensors: a comparative study. Sensors, 12(5), 5791-5814.