Commercial vehicle route planning based on dual dynamic programming and MPC algorithm

YAN Li-bing; FENG Jian-wei; XIONG Jia-wei; SHEN Zong; FENG Hai-hao; LÜ Xian-yong

doi:10.7641/CTA.2024.40048

Abstract：With the proposal of the century goal of"carbon peak and carbon neutrality",the energy consumption industry represented by commercial vehicles and passenger vehicles has begun a new round of technological revolution.With the development of autonomous driving and intelligent network technology,intelligent driving technology that integrates road information is developing rapidly.By reasonably planning vehicle speed,reducing unnecessary braking and shifting of vehicles,and reducing waiting time at traffic lights,the fuel economy of vehicles can be improved.This article aims to develop a commercial vehicle route planning algorithm with advantages such as high efficiency and economy,and proposes a control architecture based on two-layer dynamic programming+model predictive control(MPC).It combines information such as traffic lights,speed limits,and slopes on the road ahead to plan the vehicle speed and position,and splits the vehicle travel time and fuel consumption to reduce coupling,ensuring that the vehicle can further improve fuel economy while quickly passing through the road ahead.At the same time,considering the deviation in the planned path of vehicle tracking,the actual position of the vehicle is closed-looped through MPC,and the acceleration limit is taken into account to enable the vehicle to complete the tracking of the planned path,ultimately achieving energy conservation and consumption reduction for the vehicle.This article verifies the superiority of the proposed control strategy applied to commercial vehicle path planning through joint simulation using the Simulink+GT-SUITE platform.Compared with the single-layer dynamic programming scheme,this scheme can save 2.13％fuel.