Machine Learning-Based Shoveling Trajectory Optimization of Wheel Loader for Fuel Consumption Reduction

The difference in fuel consumption of wheel loaders can be more than 30% according to different shoveling trajectories for operations, and the optimization is an important way reduce operations. existing trajectory method mainly through theoretical calculation simulation analysis, which cannot fully consider high randomness complexity process. It difficult achieve desired effect. Therefore, this paper takes actual operation data as basis. factors that have a impact on are screened out Kernel Principal Component Analysis. Moreover, mathematical model established by Support Vector Machine combined with Improved Particle Swarm Optimization algorithm optimize trajectory. To demonstrate generalization ability model, two materials, gravel, sand, selected. Meanwhile, influence engine speeds considered. We three speeds. optimized verified compared sample manually controlled data. results show 27.66% 24.34% gravel respectively. This study provides guidance energy-efficient loaders.