Research on Lane Changing Scenario Analysis and Fusion Decision Model of Intelligent Material Transport Vehicle

Abstract

With the rapid development of the intelligent material transportation industry and the landing and maturity of the closed road transportation scene, the research and landing of the open road automatic driving transportation scheme has become a hot spot in the current competition of various material transportation industries. Previous studies mainly focus on the optimization of control decisions in the algorithm, which improves the performance but increases the computational cost of the system, which is not conducive to the productization of the system. This paper introduces a networked collaborative material transport architecture based on cloud, road and vehicle to reduce the cost of vehicle environment perception. Through the study of the common scene and dangerous scene of car lane change, the difficulties of automatic driving in lane change technology are analysed, and the steering structure of the four kinds of transport trolleys is compared to study the adaptability of the transport scene. Considering the cost of the main control system, this paper designs a lane change decision model that integrates the rule model and learning model, and tests 11 lane change scenarios in four types of venues in the indoor open test field of the laboratory to detect the lane change safety, decision efficiency, and lane change control smoothness. The results show that the performance of the fusion model designed in this paper does not lose the learning model, and can run on the microcontroller. Compared with the previous schemes, it saves the cost of computing power and provides a reference for the intelligent material transportation scheme of wharf, hospital and production workshop.