A Rolling Prediction-Based Multi-Scale Fusion Velocity Prediction Method Considering Road Slope Driving Characteristics

Abstract

<div class="section abstract"><div class="htmlview paragraph">Velocity prediction on hilly road can be applied to the energy-saving predictive control of intelligent vehicles. However, the existing methods do not deeply analyze the difference and diversity of road slope driving characteristics, which affects prediction performance of some prediction method. To further improve the prediction performance on road slope, and different road slope driving features are fully exploited and integrated with the common prediction method. A rolling prediction-based multi-scale fusion prediction considering road slope transition driving characteristics is proposed in this study. Amounts of driving data in hilly sections were collected by the advanced technology and equipment. The Markov chain model was used to construct the velocity and acceleration joint state transition characteristics under each road slope transition pair, which expresses the obvious driving difference characteristics when the road slope changes. An algorithm was designed to satisfy velocity continuity and boundary constraints required by road slope. Then, based on the relationship between prediction distance and weight value, using the prediction information of actual historical data, a rolling prediction-based multi-scale fusion prediction algorithm was designed to predict future velocity in the prediction horizon. Compared with the rolling prediction-based multi-scale fusion prediction without considering the road slope transition characteristics and the nonlinear neural network prediction method, the proposed method shows better prediction performance, which shows the necessity of considering different characteristics with the road slope. The verification results show that in a reasonable prediction horizon, the prediction deviation of the proposed method can be within 1km/h, and the average calculation time can be within 1s, and the prediction performance can meet the requirement of practical application, which will be helpful for studying advanced energy-saving driving assistance systems of commercial self-driving vehicles on mountainous routes.</div></div>