Wavelet neural network applied for prognostication of contact pressure between soil and driving wheel

摘要: This paper describes the measurement of contact pressure in context wheel–terrain interaction as affected by wheel load and tire inflation when fusion wavelet transform with back-propagation (BP) neural network is applied to construct prediction model. To this aim, a controlled soil bin testing facility equipped single-wheel tester was utilized while three levels velocity, slippage were applied. Using image processing technique, area values determined which subsequently used for quantification pressure. Performances different predictor models incorporated various mother wavelets evaluated using standard statistical evaluation criteria. Root mean square error coefficient determination 0.1382 0.9864 achieved optimal are better than that BP network. The proposed tool typifies high learning speed, enhanced predicting accuracy, strong robustness.