Prediction of Accelerator Pedal Position Value Based on SVR Optimized by the Hippopotamus Algorithm

The operation of advanced driver-assistance systems in modern automobiles is contingent upon a precise understanding of vehicle dynamics; an understanding of the anticipated changes of the accelerator pedal enables these systems to better adapt to real-time traffic conditions and make more precise responses. To enhance the prediction accuracy of Support Vector Regression (SVR) for the accelerator pedal position value, this paper proposes an accelerator pedal prediction model that utilizes the Hippopotamus Optimization (HO) algorithm to optimize the SVR. Initially, data from the wiring harness and peak data from the power battery are collected during the operation of a new energy vehicle, and characteristic quantities are selected from this dataset to serve as the input features. Subsequently, the HO algorithm is utilized to optimize the penalty factor C and the kernel function parameter γ of the SVR, thereby generating a prediction model with higher accuracy. Finally, the effectiveness of the proposed model was validated through comparison with other methods. The experimental results indicate that the HO-SVR model improves upon the SVR and BP models by 42.25% and 50.90%, respectively. It is thus demonstrated that the proposed model possesses practical significance for the prediction of the accelerator pedal position value.