Research on Collaborative Optimization Intelligent Driving Control System for New Energy Vehicles in Urban Traffic
Year: 2026
Type: conference-paper
Abstract: Addressing the challenges of frequent start-stop cycles, intersection queuing, energy consumption fluctuations, fragmented vehicle-road information, and lack of global coordination in single-vehicle intelligent driving control for new energy vehicles in urban traffic environments, this paper proposes a collaborative optimization intelligent driving control system (CEV-CODC) for new energy vehicles. This system adopts a three-layer structure of vehicle-side intelligent control, roadside edge collaboration, and cloud-based global optimization, uniformly encoding vehicle position, speed, acceleration, SOC, traffic light phase, road segment density, and queue length into a dynamic traffic state map. At the algorithm level, this paper constructs a multi-objective optimization function integrating travel time, energy consumption per unit mileage, speed stability, safe distance, and intersection congestion penalties, and designs a multi-agent collaborative decision-making mechanism to achieve joint control of speed, path, and energy recovery intensity of new energy vehicles in urban multi-intersection scenarios. The experimental section constructs scenarios such as multi-signal intersections, high-density congestion, low SOC constraints, and mixed traffic flow based on the SUMO concept and a Python simulation environment, and compares the results with PID, MPC, DQN, and ordinary MARL methods. Simulation results show that the proposed method can maintain a lower average travel time, a smoother SOC decline curve, smaller speed fluctuations, and a lower safe distance violation rate under high traffic density, verifying the effectiveness of the system in the intelligent driving cooperative control of urban new energy vehicles.
Authors: Qin Wanxing, Zhou Heng, Wang Zheyue, Xing Bin
Institution: Hubei Industrial Polytechnic
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7203666151