The traffic pattern at unconventional intersections involves setting up pre-signals upstream of the intersection to organize left-turn and straight-ahead traffic flows to alternate use of entry lanes, thereby increasing traffic capacity. This is a new approach to solving congestion at intersections. Although it has been practically applied in several cities in China, many times it has ended in failure due to a lack of solid theoretical and technical support. Our team will build a failure probability model for unconventional intersections based on our existing research results, establish mechanisms and optimization methods for preventing deadlock, dynamic activation, and dynamic control; in addition, based on research on driving behavior and choice preferences, we will design a dynamic and static traffic language system that comprehensively considers driver adaptability and lane queueing equilibrium. The project has already published 8 related papers in its early stages, obtained 1 national invention patent and 1 software copyright. On this basis, through cooperation with outstanding enterprises and units, With the aim of being application-oriented, on one hand, we continue to deepen the scientific and practical nature of the results. On the other hand, we achieve the promotion and application of the results, providing new technical support for research institutions, transportation planning, design, and management fields. It provides effective application systems for preventing and alleviating traffic congestion, solving actual traffic problems, while serving the improvement of the overall quality of the transportation system.