To obtain a qualitative understanding of tractive performance parameters, ride comfort, vibration control and the design of an off-road vehicle suspension system, it is essential to find the resultant force position on the wheel. To this aim, a soil bin facility assisted with a single-wheel tester was used for the synthesis of the objective parameter. Four levels of slip were induced to the wheel along with three levels of velocity and two wheel loads. The stochastic characteristic of soil–wheel interactions promoted the authors to apply two promising artificial intelligence approaches of artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) and compare the results with the statistical approach of multiple linear regression (MLR). Various structures of ANN and ANFIS tools were constructed to obtain the best representations. Two statistical performance criteria of mean squared error (MSE) and coefficient of determination (R2) were employed to assess the potential of the constructed models. In view of the employed criteria, it was divulged that the supervised ANN outperformed the ANFIS model with MSE and R2 values of 0.02615 and 0.93628, respectfully, where ANFIS model yielded MSE and R2 values equal to 0.0439 and 0.8494, respectfully.
- Soil bin