Driver behaviour modelling is essential for reducing driver workload and enhancing vehicle safety systems. To create more personalized and intelligent vehicles, efficient driver behaviour models must be integrated into vehicle dynamic systems. This paper focuses on understanding driver responses to unexpected events, with a particular emphasis on acceleration profiles due to their sensitivity to driving anomalies. Notable abnormal driving manoeuvres, such as hard braking, sharp U-turns, and sudden lane changes, often necessitate stopping the vehicle. The study employs KDE of histograms, KL divergence, ANOVA, and F-test analyses to identify and validate these findings. Additionally, time-frequency plots are utilized to predict various driving anomalies and their causes. A Kalman filter-based jerk estimation algorithm is proposed to estimate changes in jerk behaviour, aiding in the prediction of driving anomalies. The applications of this research include developing advanced driving safety systems by evaluating driving behaviour changes before and after anomalies occur.

Driving comfort assessment is a prerequisite to improve the journey experience for the drivers as well as the passengers. In this work, we proposed an advanced approach for the measurement of driving comfort in real-time. Different types of environmental features are considered along with the traditionally used Comfort Index (CI), and an Environment-specific Comfort Index (EsCI) is proposed. EsCI is also inversely proportional to the drivers' comfort level, just like CI. We also developed an android application named QDCL (Quantification of Driver Comfort Level) for overall data collection and computation of EsCI from the same. A series of driving experiments at different times of the day and different traffic conditions have been performed in Indian urban road scenarios to assess the performance of QDCL and the relevance of EsCI. We extended the work by studying the effects of different external stimuli on the computed driving comfort level. The performance of EsCI is observed to outperform the traditionally used CI (Comfort Index) in terms of accuracy for the quantification of overall driving comfort.

With the increasing number of road traffic deaths, transportation safety-related study has gained a significant amount of research interest. Tri-axial acceleration values and vehicle speed are observed to be the most efficient vehicular features to identify and measure unsafe driving events. An abnormality index has been proposed to quantify the extent of driving anomaly. The effects of different external stimuli, such as auditory, olfactory, and visual stimuli, over the proposed abnormality index have been studied, and the index is further validated by a set of experimental data collected from a total of 70 sessions of real-time driving experiments. These results can be incorporated further for developing better driving safety systems to cause the advancement of existing autonomous vehicles.

Driving error is one of the crucial contributing factors to the increasing number of traffic deaths all over the world. Both external and internal stimuli significantly affect the driving performance of individuals, irrespective of their mild, moderate, or aggressive driving styles. Continued research is being performed to increase the efficiency of vehicle safety systems and improvise existing autonomous and semi-autonomous vehicles. This paper reviews the existing state-of-the-art technologies for different types of abnormal driving detection. The review is categorized into three sections i.e., abnormal driving detection using i) vehicular features, ii) physiological features, and iii) hybrid features. Various approaches have been compared for abnormal driving detection and areas for improvement are distilled. The research gaps identified lie in the lack of i) consideration of environmental data, ii) non-invasive physiological data, and iii) comparative studies among different types of driving abnormalities.

Measurement of driving comfort is necessary, as an uncomfortable journey experience could be extremely hazardous for old people and patients. Overall driving comfort depends on several environmental contexts in addition to the driving behaviour. In this paper, we present QDCL, an android application, capable of measuring context specific driving comfort. RMS values of the tri-Axial acceleration values followed by FFT have been considered for defining driver behaviour-based comfort index (CI). For defining the context specific comfort index (CSCI), road condition and ambient temperature have been considered along with CI. International roughness index (IRI) has been considered for quantifying road anomaly, which can affect driving comfort. Deviation of ambient temperature from ideal temperature range of human body has also been considered for measuring CSCI. QDCL measures CSCI for each 20 s interval. The application provides an alarm to the driver in case the CI value exceeds the threshold for "a little uncomfortable" journey experience.