Researchers have developed an energy-saving control strategy for intelligent connected plug-in hybrid electric vehicles that incorporates driving-intention identification of the vehicle ahead. The approach is designed to help a following plug-in hybrid electric vehicle optimize speed, improve energy economy, and maintain comfort and safety by anticipating how the front vehicle is likely to move.
In connected driving environments, the behavior of one vehicle can strongly influence the performance of another. For a following hybrid electric vehicle, changes in the speed of the front vehicle affect tracking behavior, acceleration and braking decisions, passenger comfort, and energy consumption. If the following vehicle reacts only to the current speed of the front vehicle, it may miss useful clues about what the front vehicle is likely to do next.
The new study addresses this issue by fusing front-vehicle driving intention into the control of the following plug-in hybrid electric vehicle, or PHEV. According to the article, the method first classifies the driving intention of the front vehicle using an improved K-means clustering approach and a support vector machine algorithm. This gives the control system a way to distinguish different driving styles or behavioral patterns before making speed and energy-management decisions.
The next step is prediction. After the driving intention is identified, an Elman neural network predicts the next-step velocity of the front vehicle based on the current velocity and the identified intention. This prediction matters because speed planning for the following vehicle depends not only on where the front vehicle is now, but also on how its velocity is expected to change. Better prediction can help reduce unnecessary acceleration and braking, which in turn can improve both comfort and energy economy.
The researchers then formulate a multi-objective speed optimization problem. The optimization considers powertrain efficiency, vehicle comfort, tracking capability, and vehicle safety. This balance is important because an energy-saving control strategy cannot simply minimize fuel or electricity use at any cost. It must also keep a suitable following effect, avoid unsafe behavior, and deliver acceptable ride quality.
Energy flow management is handled through a power distribution strategy based on model predictive control with a clipped double Q-learning algorithm. In a plug-in hybrid electric vehicle, energy must be allocated across different sources and operating modes. The clipped double Q-learning component helps guide power distribution in a way that supports energy savings while accounting for the vehicle?s changing driving conditions.
The study?s simulation findings suggest that the proposed strategy can deliver both driving-performance and energy-economy benefits. The authors report that the method achieves a preferable following effect and comfort, while reaching 97.01% energy economy optimality compared with dynamic programming. Because dynamic programming is often used as a benchmark for optimal control, this result suggests that the proposed strategy can approach high-level energy-saving performance while remaining suitable for a predictive control framework.
For intelligent connected vehicles, the broader implication is that driving intention may be a valuable signal for energy management, not only for motion planning. By anticipating the behavior of the front vehicle, a following PHEV can plan smoother velocity trajectories and allocate power more effectively. Such strategies may help reduce energy consumption in real traffic, especially in scenarios where repeated speed changes would otherwise waste energy.
Further validation will still be needed in broader road conditions, traffic densities, vehicle platforms, and real-world connected-driving environments. Even so, the study offers a strong indication that combining driving-intention recognition, velocity prediction, multi-objective speed planning, and learning-based power distribution could improve energy-saving control for intelligent connected plug-in hybrid vehicles. As vehicles become more connected and electrified, control strategies that use information from surrounding vehicles may play an increasingly important role in sustainable transportation.
Reference
Author:
Guanying Liu a b , Shiquan Shen b , Yonggang Liu c , Yuanjian Zhang d , Yu Liu e , Zheng Chen b , Fengxiang Guo b
Title of original paper:
Energy-saving control of intelligent connected plug-in hybrid electric vehicle via fusing driving intention of front vehicle
Article link:
https://www.sciencedirect.com/science/article/pii/S2773153725000805
Journal:
Green Energy and Intelligent Transportation
DOI:
10.1016/j.geits.2025.100330
Affiliations:
a Faculty of Public Basic Education, Yunnan Open University, Kunming 650500, China
b Faculty of Transportation Engineering, Kunming University of Science and Technology, Kunming 650500, China
c State Key Laboratory of Mechanical Transmissions & School of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044, China
d Department of Aeronautical and Automotive Engineering, Loughborough University, Leicestershire LE11 3TU, UK
e China Automotive Technology and Research Center Co. Ltd., Tianjin 300300, China
Green Energy and Intelligent Transportation
Experimental study
Not applicable
Energy-saving control of intelligent connected plug-in hybrid electric vehicle via fusing driving intention of front vehicle
2-Jan-2026