In the past few decades, the use of autonomous vehicles has been increased at a pacing rate. Autonomous vehicles (AVs) are growing more popular because they alleviate traffic congestion and elevates safety. This is owing to the development of artificial intelligence techniques in numerous applications. Moreover, AVs improve fuel efficiency. As the number of vehicles increasing day by day the travel time also increases because of the traffic jam and it leads to the driver feeling sleepy or drowsy. Driver drowsiness is a serious problem that can lead to fatal accidents on the road. As a result, various studies have been conducted to develop effective driver drowsiness detection systems. This paper provides an overview of the problem of driver drowsiness and its consequences. It also presents a concise summary of the most relevant studies and methods that have been proposed for detecting driver drowsiness. Finally, a proposed method for driver drowsiness detection is discussed in detail. Driver drowsiness is a major cause of accidents on the road, especially during long drives or at night. Drowsy drivers have slower reaction times, impaired judgement, and decreased situational awareness. This can lead to accidents that result in injury or death. As a result, there has been considerable interest in developing effective driver drowsiness detection systems. Several studies have been conducted to detect driver drowsiness using various methods. Some of the most commonly used methods include monitoring the driver's eye movements, head movements, heart rate, and steering wheel movements. Machine learning algorithms have also been used to analyze these signals and determine the driver's level of alertness.One of the proposed methods for driver drowsiness detection involves using a combination of machine learning algorithms and a neural network. The system is trained using a large dataset of drivers in various driving conditions. The system is designed to detect changes in the driver's behavior and alert the driver if they are determined to be drowsy.The proposed method is designed to be non-intrusive and easy to use. It is also capable of detecting driver drowsiness in real-time, making it suitable for use in various driving scenarios. The system has been tested on a large dataset of drivers and has shown promising results, with a high accuracy rate and low false positive rate. This drowsiness of drivers can increase the frequency of accidents. Driver tiredness is one of the primary reasons for accidents. A technique for detecting tiredness using images is suggested in this study. Eye Aspect Ratio technique (EAR) is utilized for identification of the mood of the automobile driver. In this model, a raspberry pi, the pi camera, and a GPS module were employed to recognize and monitor the condition of eye shut on a continual basis in the present situation also used ultrasound sensor for obstacle detection which will help the driver while driving. This system is capable of discerning if the driver is tired or not. The system is tested successfully in different environments like micro sleep cold conditions, wear, and low light, and approximately ninety percent of the time, the model successfully experimented with improvement in drivers alertness. This prototype can be installed in front of the driver’s dashboard to detect the facial features to extract the data and alert the driver according to the situation.
An Artificial Intelligence based Prototype of Drıver Drowsıness Detectıon for Intelligent Vehicles
2023 IEEE World AI IoT Congress (AIIoT) ; 0633-0640
07.06.2023
1945108 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch