Multi Fold computational behaviour modeling of human vehicle interaction

Abstract

Driving is considered to be one of the most complex tasks because it involves a number of other activities besides just driving. Driving safely and concentrating only on driving, should be driver s main responsibility. But he or she also needs to do a number of ancillary chores at a time. For instance, using the brake, accelerator, and clutch pedals while simultaneously moving the gears as needed, controlling the steering wheel while using the controls situated on the dashboard, and so on. It is challenging for scientists and researchers to simulate realistic driving behaviour. With this goal in mind, a Smartphone sensor generated dataset of Indian drivers was constructed with driving parameters that had a significant impact on driving behavior. As a first step, we created a dataset using Smartphone sensors such as the accelerometer and gyroscope. The dataset was used in the training, testing, and validation for building a machine learning model for driver behavior classification. When driving, the most crucial factor to consider is your own and the passengers safety. Drivers must be kept under observation for any potential harmful act, whether intentional or inadvertent, in order to ensure a safe navigation of the vehicle. A real-time emotion detection system for a driver was developed to detect, exploit, and evaluate the driver s emotional state using drivers facial expressions. This study focuses on the creation of an intelligent system for face image-based expression classification using convolutional neural networks.