Fusion of adaptive cruise control algorithm and artificial intelligence model for new energy vehicles

Aiming at the performance improvement demand of adaptive cruise control for new energy vehicles in complex scenarios, this paper proposes an adaptive cruise control algorithm based on multimodal data fusion and reinforcement learning optimization. The algorithm integrates multi-source sensor data such as millimeter-wave radar and camera to build a road environment perception model, and optimizes the control strategy with safety, comfort and energy saving as the goals through reinforcement learning. On the MATLAB/Simulink and CarSim joint simulation platform, typical scenarios such as urban roads and highways and special working conditions such as emergency braking and front vehicle jamming are set for testing. The results show that compared with the traditional PID control algorithm, the new algorithm reduces the number of collisions by 68%, and the safety distance compliance rate increases from 75% to 96%; the average value of the acceleration change rate decreases by 45%, the number of rapid acceleration and deceleration decreases by 62%, and the driving comfort is significantly improved; the energy consumption is reduced by 32%, and the cruising range increases by about 28%; at the same time, the algorithm calculation time is only 0.011s, which meets the realtime control requirements. Simulation data show that the algorithm outperforms traditional algorithms in many aspects and provides a reliable solution for the intelligent development of new energy vehicles.