Today, many transportation and distribution companies entrust their vehicles to their
employees. As of 2023, the cheapest new car price in Turkey is at least 750 thousand liras.
In addition, with inflation and increasing economic costs, the cost of damage in case of an
accident has also increased. Therefore, even if the damages are covered by insurance or
car insurance, the loss of value in the vehicle is reflected in the money to the companies. In
addition to material damage, there is also loss of life in accidents. Our aim in this project is to
create driving profiles of drivers so that companies such as Trendyol, Migros, MNG Kargo
can have detailed information about the drivers. In this way, companies will be able to
evaluate their drivers' driving and see which route the vehicle is traveling on. The choice
companies make will prevent future loss of life and property.
We aim to provide users/customers with information about the driving of company
vehicles through parameters by which vehicle usage can be evaluated, such as tracking
vehicle fleets, driving analysis and driver profiling using Arduino or other embedded systems.
By using ready-made modules that assist GSM/GPRS signals and measure
acceleration/inclination/G force, we aim to obtain the necessary information about driving
from the embedded system and establish the necessary connections with the remote
computer to send this information to the data processing stage.
The received driving information is analyzed and classified by artificial intelligence
algorithms on the remote computer. The new driving analysis is reused during the artificial
intelligence algorithm training phase to be used in later analyses, thus making the online
learning cycle continuous. Analysis results are included in the database to be displayed in
the mobile application.
Past driving and instant driving information analyzed as a result of the processed
data can be followed by the user with the mobile application we will develop. More than one
vehicle and driving information can be observed on a single screen at the same time.