Pattern Recognition of Impact in Passenger Car Bumper Using Smart Materials

Abstract:

All over the world at present, injuries and fatalities from road accidents are a significant problem, especially occurred to pedestrians from passenger cars. Most of the fatal accidents between car and pedestrians prove deadly because of the head to hood impact. This paper aims to develop a bumper to avoid from passenger car’s fatal head impact for pedestrian protection, and most importantly, the impact of different objects with passenger car bumper needs to be recognized. Firstly, a pendulum system is constructed to perform the fundamental research which is concentrated on the response pattern of impact-object simulation tests, and we confirmed the application possibility for the method of discriminable pattern recognition whether impact-object is human-like or not by means of neural networks using smart PZT materials. Finally, the impact characteristics analyses can provide enough pattern recognition indices which can be developed and then used to recognize the impact information by two different neural networks.

• Periodical: Key Engineering Materials (Volumes 462-463)
• Pages: 888-893
• DOI: https://doi.org/10.4028/www.scientific.net/KEM.462-463.888
• Online since: January 2011
• Authors: Gao Ping Wang, Pan Zhou, Han Xin Chen, Jun Chen
• Keywords: Bumper, Impact, Passenger Car, Pattern Recognition, Smart Materials

References

[1] Information on http: /www. oecd. org/statisticsdata.

[2] Information on http: /www. oecd. org/statisticsdata.

[3] Information on http: /www. euroncap. com/Content-Web-Page.

[4] A Comparative Study of Accident Statistics for OECD Countries, Center for Accident Analysis, Korea Road Traffic Safety Authority (2004).

[5] Statistical Analysis on Traffic Accidents, Korea Road Traffic Safety Authority (2004).

[6] F.P. Bu and C.Y. Chao: Intelligent Vehicles Symposium, Proceedings IEEE (2005), p.100.

[7] ANSI/IEEE Std, 176-1987. IEEE Standard on Piezoelectricity.

[8] H. Demuth, M. Beale and M. Hagan: Neural networks toolbox 6 user's guide, The MathWorks, Inc. (2008).