Identification and early warning method for safety characteristic of vehicle short-distance communication network

Abstract

The invention provides an identification and early-warning method for a safety characteristic of a vehicle short-distance communication network. A vehicle node obtains heartbeat information accordingto an obtained surrounding node, the heartbeat information is processed to obtain identification characteristic data, a safety characteristic classification tree is constructed by taking the identification characteristic data as training data, multiple safety characteristic classification trees are constructed and integrated for each safety characteristic of safety early warning and form a safetycharacteristic classification tree set, multiple corresponding weak classifiers are determined, for all safety characteristics of safety early warning, the weak classifiers and weight values thereof are integrated to construct a strong classifier, and in the driving process, real-time identification characteristic data serves as input data, the strong classifier carries out safety characteristic identification and safety early warning, and the strong classifier is updated by adjusting the weights of the weak classifiers according to the early warning quality. The vehicle node collects comprehensive and accurate data, the data is classified, determined and learned continuously, and a safety early warning capability adapted to the driving environment is provided.

Description

technical field [0001] The invention relates to an identification and early warning method for safety features of a vehicle-mounted short-distance communication network, and belongs to the technical field of vehicle-mounted network communication. Background technique [0002] Vehicle short-distance communication (Vehicle to X: V2X) network is realized by short-distance communication technologies such as wireless communication, GPS / GIS, and sensors. Communication between the workshop (Vehicle-2-Vehicle), the outside of the vehicle (vehicle-2-Infrastructure), and the vehicle (Vehicle-2-Person). [0003] In the process of driving the vehicle, the driver can make judgments based on the observed environment and surrounding vehicle conditions based on experience, and take corresponding measures to avoid possible dangers, and with the accumulation of driving experience, the judgments made will become more and more accurate . With the rapid development of V2X technology, the infor...

AI Technical Summary

Problems solved by technology

Existing machine learning algorithms, such as the boosting algorithm, feature classification is relatively simple, and the classification results are not evaluated, while the safety features (such as road conditions, weather, vehicle speed, location, etc.) used by vehicle nodes for safety warnings are complex and changeable. Therefore, Applying existing machine learning algorithms to vehicle nodes cannot achieve the ideal safety warning effect

Images