Pattern recognition of vehicle types and reliability analysis of pneumatic tube test data under mixed traffic condition

Pneumatic tube traffic flow testing equipment is one of the most widely applied temporary traffic information collection instruments nowadays. The imported equipment is incapable of detecting the mixed traffic due to the single traffic environment in western countries. We developed the pressure tube traffic data acquisition and analysis equipment, which is suitable for the ubiquitous mixed traffic in China. The equipment can identify motor vehicles, motorcycles and bicycle more accurately after adjusting the classic equipment structure and algorithm. The reliability and accuracy of the equipment are analyzed in the corresponding conditions of traffic flow using mathematical statistics and analysis method of probability theory. The result indicates that the device can detect parameters of traffic flow such as traffic flow, speed, and models under mixed traffic condition, and the detecting precision meets the requirements basically, but reduces with traffic increasing.