A Machine Learning Framework for Automated Accident Detection Based on Multimodal Sensors in Cars

Sensor-Based Classification of Primary and Secondary Car Driver Activities Using Convolutional Neural Networks

To drive safely, the driver must be aware of the surroundings, pay attention to the road traffic, and be ready to adapt to new circumstances. Most studies on driving safety focus on detecting anomalies in driver behavior and monitoring cognitive capabilities in drivers. In our study, we proposed a classifier for basic activities in driving a car, based on a similar approach that could be applied to the recognition of basic activities in daily life, that is, using electrooculographic (EOG) signals and a one-dimensional convolutional neural network (1D CNN). Our classifier achieved an accuracy of 80% for the 16 primary and secondary activities. The accuracy related to activities in driving, including crossroad, parking, roundabout, and secondary activities, was 97.9%, 96.8%, 97.4%, and 99.5%, respectively. The F1 score for secondary driving actions (0.99) was higher than for primary driving activities (0.93-0.94). Furthermore, using the same algorithm, it was possible to distinguish four activities related to activities of daily life that were secondary activities when driving a car.

Analysis of the Impact of Invisible Road Icing on Selected Parameters of a Minibus Vehicle

The measurement of acceleration during vehicle motion can be used to assess the driving styles and behaviours of drivers, to control vehicle traffic, to detect uncontrolled vehicle behaviour, and to prevent accidents. The measurement of acceleration during vehicle motion on an icy road can be used to warn the driver about changing conditions and the related hazards. This paper presents the results of testing the motion parameters of a Ford Transit adapted for passenger transport in critical traffic conditions. It can contribute to the improvement of road safety. Critical traffic conditions are deemed in the paper as sudden braking, rapid acceleration, and circular vehicle motion at maximum speed maintainable in the given conditions. The vehicle's acceleration and speed were measured during the tests. The tests were carried out with a TAA linear acceleration sensor and a Correvit S-350 Aqua optoelectronic sensor. The same test runs were conducted on a dry surface, a wet (after rain) surface and a surface covered with a thin, invisible ice layer. The objective of the tests was to determine the impact of invisible road icing, the so-called black ice, on the tested vehicle's braking, acceleration, and circular motion. It was demonstrated that a virtually invisible ice layer covering the road surface has a substantial impact on the tested vehicle's motion parameters, thereby affecting traffic safety. It substantially extends the braking and acceleration distances and requires the driver to reduce the vehicle's speed when performing circular motions. A clear wet surface, representing motion after rain, did not substantially affect the analysed parameters. The obtained results can be used in traffic simulations and to analyse the causes of accidents.