Vehicle control method, system and equipment based on electroencephalogram signal and multi-mode large model and medium

By establishing a thought-action mapping dictionary in both stationary and autonomous driving states and decoding driving intention features using a multimodal large model, combined with visual evoked potential signals for fusion and conflict arbitration, the low accuracy of intention recognition and the time lag of physical takeover methods in brain-computer interface-assisted driving are solved, achieving high-precision vehicle control and smooth human-machine collaboration.

CN122443483APending Publication Date: 2026-07-24BEIJING ELECTRIC VEHICLE
0 Cites 0 Cited by

Patent Information

Application Number
CN202610804381.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-05
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

In existing technologies, brain-computer interface-assisted driving suffers from low intent recognition accuracy, leading to driver cognitive fatigue, insufficient control resolution, inconsistencies between autonomous driving decisions and human intentions, time lag in physical takeover methods, weak proactive correction capabilities, and poor smoothness of human-machine collaboration.

Method used

By establishing a thought-action mapping dictionary when the vehicle is stationary, the driving intention features are decoded in autonomous driving mode using EEG signals and a multimodal large model. Combined with visual evoked potential signals, multiple alternative adjustment operations are generated, and the final operation is determined through conflict arbitration. Error signals are analyzed in real time for correction and adjustment.

Benefits of technology

It improves the accuracy of driver intent recognition and spatial trajectory control, realizes dynamic collaboration between human and machine decision-making, enhances the vehicle's active adaptive correction capability under abnormal operating conditions, and improves driving safety and continuity.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to a vehicle control method, system and device based on electroencephalogram signals and a multi-mode large model and a medium, and the method comprises the steps: building an idea action mapping dictionary in a static state; in the automatic driving state, the second state electroencephalogram signals are decoded into driving intention features in combination with the dictionary; projecting a virtual idea aim point through a vehicle-mounted visual guidance device, and fusing the excited visual evoked potential signal with the driving intention feature to obtain a comprehensive driving intention; performing cross-modal cooperative reasoning and conflict cooperative arbitration on the real-time vehicle environment perception data and the comprehensive driving intention by using a multi-modal large model to determine a final adjustment operation; after execution, the electroencephalogram signals in the third state are analyzed, and if it is judged that error-related negative potential signals exceeding a threshold value exist, self-adaptive generation is carried out, and deviation correction adjustment operation is executed. Therefore, the cognitive load of a driver is relieved, the track control precision is improved, and the man-machine cooperation smoothness and the driving safety are improved.
Need to check novelty before this filing date? Find Prior Art