Abstract
Due to the development of leisure sports industry and the increase in delivery demand, the demand for two-wheeled vehicles such as the motorcycle is increasing every year; moreover, the motorcycle accident rate is increasing. The motorcyclist’s body is exposed to the outside, and in cases of accidents, the head and the neck are particularly vulnerable. This paper proposes a study about an air bag equipped with Artificial Intelligence to protect the driver’s neck spine from motorcycle accidents. Through the six-axis sensor, it receives the driver’s motor condition data about the acceleration and angular velocity data and measures real time speed and angle; combines them with algorithms that can judge accidents through Artificial Intelligence learning to activate airbags in real time. Data were collected and learned by dividing the types of accidents; for Artificial Intelligence learning, the general Neural Network method was not used however, a mix of parallel Neural Network with an existing Neural Network were used instead. The Artificial Intelligence learning method proposed in this paper has been found to have more improved accuracy, stability and learning time compared to the existing Neural Network.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig2_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig4_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig5_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig7_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig8_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig9_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig10_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig11_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig12_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig13_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig14_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig15_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig16_HTML.jpg)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs42835-019-00229-3/MediaObjects/42835_2019_229_Fig17_HTML.jpg)
Similar content being viewed by others
References
KOSHA http://www.kosha.or.kr/trList.do?medSeq=37830&codeSeq=1150160&medForm=101&menuId=-1150160101. Accessed Mar 30 2017
Kiguchi K, Matsuo R (2017) Accident prediction based on motion data for perception-assist with a power-assist robot. In: Computational intelligence (SSCI), 2017 IEEE symposium series on (pp 1–5). IEEE
Kawaguchi S, Takemura H, Mizoguchi H, Kusunoki F, Egusa R, Funaoi H, Sugimoto M (2017) Accuracy evaluation of hand motion measurement using 3D range image sensor. In: Sensing technology (ICST), 2017 eleventh international conference on (pp 1–4). IEEE
Stone EE, Skubic M (2015) Fall detection in homes of older adults using the microsoft kinect. IEEE J Biomed Health Inform 19(1):290–301
Ryu JT (2013) The development of fall detection system using 3-axis acceleration sensor and tilt sensor. J Korea Ind Inform Syst Res 18(4):19–24
Kim SH, Park **, Kim D-W, Kim N-G (2011) The study of realtime fall detection system with accelerometer and tilt sensor. J Korean Soc Precis Eng 28(11):1330–1338
Kim H, Min JK (2011) Implementation of a motion capture system using 3-axis accelerometer. J KIISE 17(6):383–388
Lee S-M, Jo H-R, Yoon S-M (2016) Machine learning analysis for human behavior recognition based on 3-axis acceleration sensor. J Korean Inst Commun Sci 33(10):65–70
Lee H, Lee S (2014) Real-time activity and posture recognition with combined acceleration sensor data from smartphone and wearable device. J KISS 41(8):586–597
Kau LJ, Chen CS (2015) A smart phone-based pocket fall accident detection, positioning, and rescue system. IEEE J Biomed Health Inform 19(1):44–56
Gam JH, Jo SH, Woo J, Kim DW, Byun GS, Cho HR, Kim JH (2018) A study on motion recognition using 9-axis IMU sensor module. In: The 2018 spring conference of intelligent transport system, 1030–1031
Yoo B-H, Heo G (2017) Detection of rotations in jump rope using complementary filter. J Korea Inst Inform Commun Eng 21(1):8–16
Zhang Q et al (2014) Design and realization of a wearable hip-airbag system for fall protection. Appl Mech Mater 461:667–674
Sugomori Y (2017) Detailed deep learning—time series data processing by TensorFlow · Keras, Wikibook, chap.3.2.1, 3.3.1
Medium https://medium.com/@kmkgabia/ml-sigmoid-대신-relu-상황에-맞는-활성화-함수-사용하기-c65f620ad6fd. Accessed 26 Aug 2018
Machine Learning blog http://nmhkahn.github.io/NN. Accessed 30 Jan 2016
Kingma DP, Ba JL (2015) Adam: a Method for Stochastic Optimization. In: International conference on learning representations, 1–13
Heusel M, Ramsauer H, Unterthiner T, Nessler B, Hochreiter S (2017) GANs trained by a two time-scale update rule converge to a local nash equilibrium. In: Advances in neural information processing systems, pp 6626–6637
BEOMSU KIM’S BLOG http://shuuki4.github.io/deep%20learning/2016/05/20/Gradient-Descent-Algorithm-Overview.html. Accessed 20 May 2016
Tieleman T, Hinton G (2012) Lecture 6.5—RMSProp, COURSERA: neural networks for machine learning. Technical report
Sutskever I, Martens J, Dahl G, Hinton G (2013) On the importance of initialization and momentum in deep learning. In: Proceedings of the 30th international conference on machine learning (ICML-13), pp 1139–1147
Clevert DA, Unterthiner T, Hochreiter S (2015) Fast and accurate deep network learning by exponential linear units (elus). ar**v preprint ar**v:1511.07289
Hong JH, Han KS, Kim K (2017) Walking pattern analysis using an acceleration sensor device. J Electr Eng Technol 12(1):396–401
Yoon JW, Noh YS, Kwon YS, Kim WK, Yoon HR (2014) Improvement of dynamic respiration monitoring through sensor fusion of accelerometer and gyro-sensor. J Electr Eng Technol 9(1):334–343
Avci D (2016) An automatic diagnosis system for hepatitis diseases based on genetic wavelet kernel extreme learning machine. J Electr Eng Technol 11(4):993–1002
Wang D, Oh SK, Kim EH (2018) Design of space search-optimized polynomial neural networks with the aid of ranking selection and L2-norm regularization. J Electr Eng Technol 13(4):1724–1731
Acknowledgements
This work was supported by the KIEE.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Jo, SH., Woo, J., Jeong, JH. et al. Safety Air Bag System for Motorcycle Using Parallel Neural Networks. J. Electr. Eng. Technol. 14, 2191–2203 (2019). https://doi.org/10.1007/s42835-019-00229-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s42835-019-00229-3