A kind of intelligent driving behavior analysis method based on mobile phone sensor
Technical field
The present invention relates to driving behavior Intellectual Analysis Technology field, specifically a kind of intelligent driving behavior analysis method based on mobile phone sensor.
Background technology
Along with the development of vehicle-mounted OBD equipment and internet, the technology realization rate of car networking gets more and more, OBD from the beginning collects driving data and analyzes, and mobile phone sensor is finally collected driving data and analyzed, and the pattern of car networking is more and more diversified.Vehicle insurance UBI is the most representative, and vehicle insurance UBI, by the analysis to driving behavior, can carry out category division to vehicle insurance user, provide the foundation to the premium classification in future.Traditional OBD mobile unit can obtain driving behavior data accurately, but the equipment of rear dress exists clutter, can affect the accuracy of data, and user does not approve for this form of installation OBD equipment.Therefore, mobile phone sensor becomes driving behavior collecting device popular now, by mobile phone being fixed the object that can reach data acquisition onboard equally, also can not produce any impact to vehicle simultaneously.
Traditional OBD equipment can only obtain three anxious data (emergency brake, anxious oil, racing) and some vehicle status data of driving.And a large amount of mobile phone algorithms is the three anxious data going out to drive by obtaining mobile phone sensor data analysis too, these data are marked as calculated factor to driving behavior, but driving behavior is still perfect not as the scoring factor only according to three anxious data.
Summary of the invention
The object of the present invention is to provide a kind ofly to analyze rationally, the intelligent driving behavior analysis method based on mobile phone sensor that accuracy is high, to solve the problem proposed in above-mentioned background technology.
For achieving the above object, the invention provides following technical scheme:
Based on an intelligent driving behavior analysis method for mobile phone sensor, concrete steps are as follows:
(1) data acquisition: after APP front end is opened and recorded, APP is responsible for front end the collection of collecting mobile phone sensor and gps data, and wherein a gps data is read every a second in APP front end, every 200 milliseconds of reading primary transducer data;
(2) data prediction: pre-service is carried out to the data gathered in APP front end, is screened out by useless gps data, sensing data is once done filtering by a second;
(3) data transmission: packing after pretreated data correlation is sent to APP backstage by APP front end;
(4) data reprocessing: APP backstage is after receiving the data imported into APP front end, and resolution data bag, to try again filtering to sensing data, rejects the data exceeding thresholding;
(5) Output rusults: the data after reprocessing are carried out centralized calculation, analyzes driving behavior result, and result is back to APP front end be used for display and store.
As the present invention's further scheme: described centralized calculation is the data by acceleration, gyroscope, magnetometer and GPS, combination model storehouse analyzes driving behavior result.
Compared with prior art, the invention has the beneficial effects as follows:
The present invention adopts driving behavior Model Matching algorithm, while analyzing three anxious data, also will exceed the speed limit as the scoring factor, and classified by the driving habits of all strokes to driver of history, add the accuracy that driving behavior judges, insurance company objectively judges by the driving behavior of judged result of the present invention to driver, thus more scientific carries out classification to premium.
Accompanying drawing explanation
Fig. 1 is schematic flow sheet of the present invention.
Embodiment
Be described in more detail below in conjunction with the technical scheme of embodiment to this patent.
Refer to Fig. 1, a kind of intelligent driving behavior analysis method based on mobile phone sensor, concrete steps are as follows:
(1) data acquisition: after APP front end is opened and recorded, APP is responsible for front end the collection of collecting mobile phone sensor and gps data, and wherein a gps data is read every a second in APP front end, every 200 milliseconds of reading primary transducer data;
(2) data prediction: pre-service is carried out to the data gathered in APP front end, is screened out by useless gps data, sensing data is once done filtering by a second;
(3) data transmission: packing after pretreated data correlation is sent to APP backstage by APP front end;
(4) data reprocessing: APP backstage is after receiving the data imported into APP front end, and resolution data bag, to try again filtering to sensing data, rejects the data exceeding thresholding;
(5) Output rusults: the data after reprocessing are carried out centralized calculation, by the data of acceleration, gyroscope, magnetometer and GPS, combination model storehouse, analyzes driving behavior result, and result is back to APP front end be used for display and store.
The present invention is by reading mobile phone sensor data in real time, and to data analysis, draw three of driving anxious and hypervelocity data, by to the process of data for insurance company provides driver driving behaviour classification information, in the present invention, APP front end primary responsibility collects the collection of mobile phone sensor and gps data, and examination and pre-service are carried out to data, pretreated result will input to APP backstage algorithm, return to APP front end for display by APP backstage algorithm after calculating result.The present invention comprises the examination of APP front end data acquisition, pre-service and APP backstage driving behavior analysis algorithm, the validity of the examination of APP front end data acquisition mainly examination GPS, due to GPS search for time or poor signal time, gps data depart from scope for hundreds of rice can be reached, have a strong impact on the reliability of data, therefore, when reading gps data, invalid gps data is fallen in conjunction with the accuracy rating in GPS state and mistiming examination in APP front end, and export to APP background service after gps data and sensing data having been carried out mating, the data that APP imports into are mainly input to the driving behavior model bank in algorithm by APP backstage driving behavior analysis algorithm, pass through Model Matching, analyze various driving behavior, by drawing hypervelocity data with mating of road data, and result is returned to the display of APP front end.
The present invention adopts driving behavior Model Matching algorithm, while analyzing three anxious data, also will exceed the speed limit as the scoring factor, and classified by the driving habits of all strokes to driver of history, add the accuracy that driving behavior judges, insurance company objectively judges by the driving behavior of judged result of the present invention to driver, thus more scientific carries out classification to premium.
Above the better embodiment of this patent is explained in detail, but this patent is not limited to above-mentioned embodiment, in the ken that one skilled in the relevant art possesses, various change can also be made under the prerequisite not departing from this patent aim.