A data processing algorithm based on vehicle weigh-in-motion systems

Abstract

According to the output value of gravitational sensors and speed of vehicles, one back-propagation (BP) neural network model is established. The genetic algorithm is used to optimize the BP neural network. This method can speed up the convergence and avoid getting stuck in the local minimum. The experiment results show that the optimizing BP neural network algorithm based on genetic algorithm can reduce the average error of the calculation and prediction. And the accuracy and efficiency of the weigh-in-motion (WIM) system are improved.

Multiple optimized support vector regression for multi-sensor data fusion of weigh-in-motion system

“…[11][12][13] Multi-sensor weigh-inmotion (MS-WIM) method has been proposed for reducing this negative impact, while the weighing accuracy will be affected by various data fusion algorithms. Arithmetic averaging method is widely used in MS-WIM system, [14][15][16] but large weighing errors are easily generated due to the uncertainty of measurement environment, thereby leading to the low reliability of weighing results. Artificial neural network (ANN) algorithm, using training samples to construct input-output mapping model, is able to avoid complicated mathematical modeling process and realize approximation of nonlinear function with arbitrary precision.…”

Tire–Pavement Contact-Aware Weight Estimation for Multi-Sensor WIM Systems

“…Glover et al [14] arranged nine strip sensors at regular intervals. The static weight of the vehicle is estimated by the mean value, the median value, or the average of the highest and lowest values obtained by the sensors [15]. David Cebon et al [16] deployed 96 strip WIM sensors in Indiana and averaged the outputs of all the sensors to estimate the static weight.…”