Vehicle use identification method based on in transit data

The invention provides a vehicle use identification method based on in-transit data, which relates to the field of motion track research and calculation. The method comprises the following steps: S1,describing the daily driving track of each vehicle; S2, carrying out time sequence clustering according to the daily driving track of each vehicle, and finding out the main driving tracks of the typeof vehicles according to the clustering result; S3, calculating the proportion of different track categories to which each vehicle belongs in all the tracks, accumulating the sorted proportions in turn from large to small, and taking the track of which the accumulated value is greater than and closest to the threshold as the main driving track of each vehicle; and S4, defining the use of each vehicle according to the main driving track. According to the invention, clustering is carried out by taking the daily driving laws of vehicles as intermediate variables, the main driving laws of vehiclesare analyzed based on statistical knowledge, and the motion tracks of vehicles are described from three dimensions, namely, time, geographical location and mileage. Compared with the prior art in which vehicles themselves are clustered, the method is more in line with the actual vehicle use.