Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS

Тел.: +7 (495) 625-9241


Algorithms for evaluating some characteristics of driving style and condition of the car by its telemetry

DOI 10.18127/j19997493-201902-07


G.S. Ivanova – Dr.Sc.(Eng.), Professor, Department «Computer systems and networks», Bauman Moscow State Technical University
A.A. Evtukhov – Master, Department «Computer systems and networks», Bauman Moscow State Technical University

One of the ways to reduce the number of accidents on the roads is to create automated systems for assessing the driving style and tech-nical condition of the car. Such systems can not only give advice to drivers to improve driving style, but also, being integrated into the general system, transmit information to the law enforcement agencies about the dangerous behavior of a car while driving.
The initial data for obtaining the required characteristics can be the data obtained from the vehicle diagnostic systems. In particular, as a result of processing this data, it can be determined: confidence in the driver’s handlebars, delays in the braking system, braking aggres-siveness. For each of the listed characteristics, the article proposed a criterion and an estimation algorithm using the telemetric data that the system receives from the vehicle.
To assess the aggressiveness of braking data on the movement of the brake pedal were investigated. The sharpness of pressing a pedal is determined by the acceleration of its movement, which can be characterized by the tangent of the angle between the horizontal and a fragment of the brake pedal position graph.
To obtain a specific assessment value, it is necessary to determine the derivative of the pedal movement function. However, the func-tions obtained as a result of processing telemetry data are stepwise, which does not allow to obtain a derivative suitable for further anal-ysis. To solve this problem, it is proposed to use approximation by splines.
Analysis of the derivative allowed us to identify the areas where the inhibition is fixed. After that, the system analyzes each found area, determining the angle of the triangle built on the points of the beginning and end of the section.
To estimate the response time of the brake system, the time delay between pressing the pedal and the moment when the car starts to slow down is determined. To do this, it is necessary to take the derivative of the functions of pedal dependence and speed versus time. Taking into account the nature of the function, a linear approximation is first performed, then smoothing, and only then a derivative is taken. The resulting derivative is smoothed again to obtain a smoother graph of the function.
The most important parameter that determines the safety of driving a car is consistency and accuracy in handling the steering wheel. In this case, the oscillating movements of the steering wheel may indicate a malfunction of the car or a lack of driving experience with the driver. As a characteristic of the steering possession, it was proposed to use the «volume» of the rudders per unit of time. To obtain a characteristic based on vehicle telemetry, it is proposed to take measurements of the rudder movement function, determine the main trend in a noisy signal using a threshold value, exclude the rudder movement required to perform turns, and then calculate the required characteristic.

  1. Statistika DTP. Pravovaya zashchita avtolyubitelya. URL = (data obrashcheniya: 31.01.2019).
  2. Statistika DTP. Vavilon. URL = (data obrashcheniya: 31.01.2019).
  3. Statistika i spros na novye avtomobili v Rossii za 2018 g. Avto versiti. URL = (data obrashcheniya: 31.01.2019).
  4. Infografika. Avtostat. URL = (data obrashcheniya: 31.01.2019).
  5. Legkovye avtomobili Volvo. Volvo. URL= (data obrashcheniya: 10.01.2019).
  6. Legkovye avtomobili Mersedes. Mersedes. URL = (data obrashcheniya: 25.01.2019).
  7. Kompyuternaya diagnostika avtomobilya. Vikipediya. URL = (data obrashcheniya: 28.01.2019).
  8. Kody oshibok OBD. Bestskan. URL = (data obrashcheniya: 28.01.2019).
  9. Ellison A. Profilirovanie riskovannogo povedeniya voditelei v otnoshenii vsekh uchastnikov dorozhnogo dvizheniya. Avstralaziiskii kolledzh bezopasnosti dorozhnogo dvizheniya. URL = (data obrashcheniya: 28.01.2019).
  10. Papantoniu P. Faktory riska, povedenie voditelya i veroyatnost neschastnogo sluchaya. Afinskii natsionalnyi tekhnicheskii universitet. URL= (data obrashcheniya: 28.01.2019).
  11. Venshu V. Modelirovanie i raspoznavanie povedeniya voditelya na osnove dannykh vozhdeniya. Khindavi. URL = (data obrashcheniya: 28.01.2019).
  12. Sango Ch. Analiz i klassifikatsiya povedeniya voditelya s ispolzovaniem bortovoi informatsii CAN-BUS. Reserchgeit URL = (data obrashcheniya: 28.01.2019).
  13. Karmona Yu. Sliyanie dannykh dlya analiza povedeniya voditelya. Natsionalnaya biblioteka meditsiny SShA. URL: (data obrashcheniya: 28.01.2019).
  14. Chan F. Identifikatsiya voditelya po stilyu vozhdeniya. Stenfordskii Universitet SShA. URL: (data obrashcheniya: 28.01.2019).
  15. Ivanova G.S., Evtukhov A.A. Sistema parametrov dlya otsenki stilya vozhdeniya po telemetrii avtomobilya. Tekhnologii inzhenernykh i informatsionnykh sistem. 2019. № 1. S. 27.
  16. Diagramma tormozheniya. Khelpkins. URL = (data obrashcheniya: 25.01.2019).

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio