Vehicle running state recognition method, system and device based on user behavior data

Abstract

The invention discloses an automobile running state identification method, system and equipment based on user behavior data. The method comprises the following steps: collecting a plurality of types of user historical behavior data; Constructing a DeepConvLSTM model for each kind of user historical behavior data, and training a corresponding model by using each kind of user historical behavior data and a known automobile running state tag; training a corresponding DeepConvLSTM model for each kind of user historical behavior data; Collecting current behavior data of a user, inputting each kindof current behavior data of the user into the corresponding trained DeepConvLSTM model, outputting an automobile running state classification probability, and selecting an automobile running state corresponding to the maximum probability value as an output value of the current DeepConvLSTM model; And finally, selecting one of the output values of all the DeepConvLSTM models with the maximum numberof the automobile running states as the current automobile running state.

Description

technical field [0001] The present disclosure relates to a method, a system and a device for identifying a running state of a vehicle based on user behavior data. Background technique [0002] The statements in this section merely mention background art related to the present disclosure and do not necessarily constitute prior art. [0003] With the development of society and the improvement of people's living standards, cars are gradually and widely integrated into our lives. The appearance of automobiles has brought a lot of convenience to our daily life. However, as the number of automobiles increases, traffic congestion becomes more and more frequent. Traffic congestion increases people's travel time, which affects people's work and life; when cars are in a congested state, it will also cause unnecessary waste of fuel, making the air pollution worse; when people encounter emergencies, they need to travel It is also very likely to cause time delays due to traffic congest...

AI Technical Summary

Problems solved by technology

Traffic congestion increases people's travel time, which affects people's work and life; when cars are in a congested state, it will also cause unnecessary waste of fuel, making the air pollution worse; when people encounter emergencies, they need to travel It is also very likely to cause time delays due to traffic congestion, which brings great inconvenience to people's lives.

[0004] At present, the identification of traffic jams is mainly realized by the feature recognition of road traffic images through deep learning models. Since this method needs to set up many camera devices to collect road traffic images in real time, this method is mainly related to existing road traffic images. The combination of monitoring cameras requires additional monitoring probes on important roads in the city to achieve real-time detection of road traffic conditions, but this method is still difficult to achieve full coverage monitoring of traffic roads, and road congestion or traffic accidents occur in monitoring blind spots When the time comes, the information cannot be fed back in time, which will easily lead to the aggravation of congestion or the occurrence of secondary accidents.

Images