关键词: 交通拥堵辅助系统；识别能力；无人驾驶；蛇形修正；策略优化

Abstract: Traffic jam assist (TJA) system is one kind of intelligent driving assistance system, which can provide vertical and horizontal assistance for drivers in the case of traffic jams, and pave the way for future advanced unmanned driving.Now, in view of the situation that the matching system model has followed the preceding vehicle through the intersection scenario, and the steering is corrected in a serpentine shape or a sharp direction due to recognition delay after the vehicle enters the lane, this paper starts from the three aspects of factor analysis, strategy optimization and real vehicle verification, to eliminate the occurrence of such problems in the matching process, improve the safety and reliability of subsequent matching system models, and provide a certain design basis for subsequent driverless solutions, which will have a certain impact on the improvement of model brand power.

Key words: Traffic jam assist system; Identification ability; Driverless; Serpentine correction; Strategy optimization