The future of transportation is driven by the use of artificial intelligence to improve living and transportation. This paper presents a neural network-based system for driver identification using data collected by a smartphone. This system identifies the driver automatically, reliably and in real-time without the need for facial recognition and also does not violate privacy. The system architecture consists of three modules data collection, preprocessing and identification. In the data collection module, the data of the accelerometer and gyroscope sensors are collected using a smartphone. The preprocessing module includes noise removal, data cleaning, and segmentation. In this module, lost values will be retrieved and data of stopped vehicle will be deleted. Finally, effective statistical properties are extracted from data-windows. In the identification module, machine learning algorithms are used to identify drivers' patterns. According to experiments, the best algorithm for driver identification is MLP with a maximum accuracy of 96%. This solution can be used in future transportation to develop driver-based insurance systems as well as the development of systems used to apply penalties and incentives.
Artificial intelligence and data mining are two important strategies for the development of future transportation systems. Given the spread of traffic data, especially in the field of sensors, automobiles, IoT, telecommunications tools and smartphones, more attention is needed to these two strategies in transportation. Applications like connected vehicle control, traffic prediction according to big-data, and online decision-making in advanced driverassistance systems and smart. The purpose of this article is to use artificial intelligence to identify a driver from a set of drivers based on smartphone data. This system allows us to identify the driver using driving characteristics such as physical or behavioral characteristics. One example implemented in this regard is seen in [1].
Driver identification has a variety of applications, including intercity public transport, freight transportation, and quality driving control, as well as it can be used for special insurance policies, the imposition of fines, and in-service program [2]. For driving safety, it can help detect abnormal driving behaviors in driving evaluation systems, to detect dangerous driving or under the influence of alcohol and drugs and also to detect distraction. Driver identification is also used in intelligent transportation fleet management systems to control the amount of driving per day and avoid using an unauthorized driver. Also in the insurance industry, intelligence insurance requires driver identification and evaluation to get the price appropriate to the level of driving of car owners. For example, some companies consider high speed, midnight driving, and braking to be dangerous [3][4][5][6].
In Iran, since 2020, third party insurance will be driver-driven, so that the driver will be insured instead of the car. Factors affecting the cost of insurance include female or male, young or old, high-risk and low-risk drivers. Driver identification in the insurance industry is used to identify the unauthorized driver and the driver who has an accident. Driver identification is used in advanced driver-assistance systems to provide targeted services, for example by identifying a member of the family who is driving, it adjusts the vehicle or route settings to suit that individual [7][8][9]. Alarm systems can also be equipped with more advanced tools using machine learning [10]. Today, alarm systems use GPS as a monitoring tool. However, these systems do not automatically detect theft, but using driver identification can be equipped with automatic detection.
On the other hand, the use of IoT in driver identification is very important. It is projected that by the year 2020, 20 billion Internet of Things devices will be used in the world. These days, cars are equipped with a variety of sensors to provide advanced driver-assistance systems such as adaptive navigation control, auto park and automatic emergency brake. These sensors in vehicles lead to advances in transportation research. One of the most popular research topics is the study of driving style to identify driver behavior. A review of research into the study of driving style and the role of IoT is discussed in [11][12].
In the past, drivers were identified using simpler methods such as ID cards or notes, later more sophisticated methods such as fingerprint, face recognition as well as innovative methods such as finger vein pattern recognition was replaced [13][14][15]. However, it is easier and safer to identify a driver using a sensor because in previous systems it was possible to cheat. For example, a driver could be replaced with an unauthorized driver after moving, but the sensors reduce the possibility of fraud due to continuous evaluation. Also, since the sensors are automatic, the driver does not need any extra action. On the other hand, the use of voice and face recognition tools violates driver privacy, so the use of sensors that do not violate driver privacy is very important [3][4][5]. In this paper, a tentative basis for appropriate policy discussion is provided to balance tool and privacy in vehicle-data sharing scenarios. To this end, the potential for privacy breach in collecting vehicle sensor data is investigated [16].
Here are some of the major sensors used in research based on hidden sensors. Inertial Measurement Unit (IMU) includes accelerometer and gyroscope sensors. Smartphone or OBD-II device can be used to collect the data. The accelerometer sensor converts the mechanical acceleration into an electrical signal accordingly. Acceleration is the rate of change of velocity of an object concerning time. The accelerometer has single-axis and multiaxis models that can measure the size and orientation of the acceleration as a vector. Accelerometers sense the forces that cause acceleration (both static and dynamic). An example of a static force is the force of gravity And for dynamic forces, vibrations can be mentioned. A gyroscope is a device used for measuring or m
This content is AI-processed based on open access ArXiv data.
