汽车碰撞试验速度的实时等误差测量和神经网络修正

The Equal Error Measurement and Neural Network Correction for Real Time Velocity in Car Crash Test

汽车碰撞试验已经成为强制性法规试验检测项目,速度的准确控制和测量是试验的关键环节之一。文章在分析传统汽车速度测量方法、误差和局限性的基础上,提出了试验过程实时测量和闭环控制的方法,可以在不提高系统传感器要求的条件下实现车速的等精度测量,并且应用神经网络方法,对测量中的随机误差等进行实时修正,从而获得很好的测量效果。所得结论可为设计相关测速系统和提高测量准确度提供参考。

Car crash test is one of the tests in vehicle design regulations in China, and the precise measurement and control of vehicle speed are the key factor of it. Based on the analysis of such measurement theory, measurement error and localization, a new method is put forward, which can be used in the real time measurement and the closed loop control. This method can also realize the equal error measurement for vehicle speed without promoting the specification of sensors. The random error in the process of measurement is also corrected by the method of neural network, and then a good result can be obtained. These conclusions can be used as references for speed measurement and control system design and increase the accuracy .