Traffic route dynamic induction technology based on coupling of time recursion and neural network

Abstract

The invention discloses a traffic route dynamic induction technology based on coupling of time recursion and neural network. The traffic route dynamic induction technology based on the coupling of the time recursion and the neural network is characterized in that based on the coupling relationship of 'man-machine-environment', the relationship between road conditions and time variation is studied. Difference of road conditions and human factor actions in different times and the randomness of unexpected road accidents are synthesized, neural network learning with a teacher is taken as an experience accumulating method, a time recursion predicting method is put forward, the shortest time of route passing is determined, and therefore dynamic induction of a traffic route can be achieved. Recursion prediction is conducted through comparison between accumulated experience of a knowledge database and information about real-time road condition, and therefore real-time and effective road condition information support is provided for a driver. According to the traffic route dynamic induction technology based on the coupling of the time recursion and the neural network, the overall logic structure comprises a knowledge base structure, path information learning with the teacher, human factor influence on route selection and logic core model establishment. The traffic route dynamic induction technology based on the coupling of the recursion and the neural network can assist the driver in making a correct judgment on the road condition in time, reduces time loss caused by insufficient experience and unexpected accidents, and can be widely used for traffic path dynamic induction of vehicles.

Description

technical field [0001] The invention relates to the field of urban traffic guidance, in particular to the dynamic guidance of traffic paths based on time recursion and neural network coupling. Background technique [0002] Urban traffic congestion is largely due to the lack of predictability and guidance of the obtained road condition information. Drivers only choose routes based on experience and real-time road conditions, which often makes the distribution of traffic volume on the road network uneven. How to develop a scientific and effective path guidance technology to adapt to the ever-changing traffic conditions and adjust the driving route in real time so that the driver can minimize the useless traffic waiting time in the crowded urban traffic environment is particularly important. [0003] Over the years, experts and scholars have proposed many traffic route search models and algorithms to solve the route selection problem. The more classic ones are the Dijkstra mod...

AI Technical Summary

Problems solved by technology

[0004] However, the existing path guidance technology seldom considers the continuous change characteristics of time. The formulation of the guidance strategy usually selects the instant state of the entire path at any point in a certain road section, and fails to realize In the continuous period of time, the cumulative experience time of the entire path is compared with the real-time path situation, so that the calculation results have a large deviation from the facts; it is difficult to estimate the traffic status of the remaining paths based on experience at a point in time; without considering The impact of emergencies on the road and avoidance; only consider the influence factors of machine and environment, and seldom consider the influence of the essential characteristics of people as the main body of traffic

Images