Search

Monitoring drivers' mental workload facilitates initiating and maintaining safe interactions with in-vehicle information systems, and thus delivers adaptive human machine interaction with reduced impact on the primary task of driving. In this paper, we tackle the problem of workload estimation from driving performance data. First, we present a novel on-road study for collecting subjective workload data via a modified peripheral detection task in naturalistic settings. Key environmental factors that induce a high mental workload are identified via video analysis, e.g. junctions and behaviour of vehicle in front. Second, a supervised learning framework using state-of-the-art time series classifiers (e.g. convolutional neural network and transform techniques) is introduced to profile drivers based on the average workload they experience during a journey. A Bayesian filtering approach is then proposed for sequentially estimating, in (near) real-time, the driver's instantaneous workload. This computationally efficient and flexible method can be easily personalised to a driver (e.g. incorporate their inferred average workload profile), adapted to driving/environmental contexts (e.g. road type) and extended with data streams from new sources. The efficacy of the presented profiling and instantaneous workload estimation approaches are demonstrated using the on-road study data, showing $F_{1}$ scores of up to 92% and 81%, respectively.

In the driving environment, competition exists between Driving Related Activities (DRAs) and Non-Driving Related Activities (NDRAs). This is a source of inattention and human error. Continual proliferation of in-vehicle information systems (IVIS) presents drivers with opportunities for distraction. Drivers simultaneously manage DRAs alongside unrelated but cognitively demanding NDRAs. Vehicle designers need ways of understanding human capability in such situations to provide solutions that accommodate these conflicting demands. This paper proposes a framework intended to address such challenges, rooted in the widely accepted construct of Situation Awareness (SA). However, SA theory does not presently accommodate disparate unrelated goal-driven tasks performed in parallel. This framework reconciles the present reality of drivers simultaneously devoting cognitive resources to attain SA for multiple activities by proposing a separate body of knowledge for each active goal. Additionally, the process of achieving SA is expanded to incorporate this concurrent development of separate bodies of goal-directed knowledge. The advantage of reconceptualising SA for driving allows consideration of interface design which minimises the impact of competing activities. The aim is a framework facilitating creation of IVIS that help drivers succeed in multi-goal multitasking situations. Implications of the proposed framework for theory, design, and industry-driven automotive safety efforts are discussed.

The modern-day vehicle hosts a complex environment where competing goals challenge the act of driving. Drivers may pursue Non-Driving Related Activities (NDRAs), resulting in multitasking. How NDRAs are achieved will depend upon whether In-Vehicle Information Systems (IVIS) are available. If not, alternative means may be taken that could compromise Driving Related Activity (DRA) performance. Situation Awareness (SA) is central to achieving multiple goals in a dynamic environment. This thesis investigates an Unconstrained approach to designing IVIS to examine how to improve multitasking performance in-vehicle through enhanced driver SA. A combination of theoretical and experimental activities were performed. A literature review in SA theory exposed a lack of detailed consideration for competing goals and task switching. A restated model of SA was proposed to show the impact on Situation Assessment and the Situation Model of the act of multitasking. Using a new methodological approach, a series of simulator experiments then aimed to measure and characterise the effect of increased awareness on task performance when multitasking. The results demonstrated that increased awareness improved task performance, therefore, providing scope for IVIS to achieve the same. Subsequently, by using the model proposed, IVIS concepts were designed to support either the NDRA or DRA, and a simulator experiment was performed. Supporting NDRAs using Contextual Cuing proved successful by reducing the cost of task switching, but impacted overtaking behaviour. Supporting the DRA, while successful for Lane-keeping, did not operate as intended. Drivers were not able to adequately use concurrent feedback due to not achieving a skilled level of behaviour. Finally, increased awareness impacted how a driver manages their attention. Drivers acted more strategically, showing evidence for dynamic compensatory metacognition. Therefore, by using SA to understand the vehicle environment, in detail, and taking an Unconstrained approach to IVIS design, improved in-vehicle multitasking performance can be achieved., This project was funded by Jaguar Land Rover Limited

Technological and sociological developments in the automotive sector are shifting the focus of design towards developing a better understanding of driver needs, desires and emotions. Human centred design methods are being more frequently applied to automotive research, including the use of systems to detect human emotions in real-time. One method for a non-contact measurement of emotion with low intrusiveness is Facial-Expression Analysis (FEA). This paper describes a research study investigating emotional responses of 22 participants in a naturalistic driving environment by applying a multi-method approach. The research explored the possibility to investigate emotional responses and their frequencies during naturalistic driving through real-time FEA. Observational analysis was conducted to assign causes to the collected emotional responses. In total, 730 emotional responses were measured in the collective study time of 440 minutes. Causes were assigned to 92% of the measured emotional responses. This research establishes and validates a methodology for the study of emotions and their causes in the driving environment through which systems and factors causing positive and negative emotional effects can be identified.

Selection and comparison of human-factors related measures for evaluations of in-vehicle\ud devices involves weighting of multiple criteria. It may result in a complex decision-making process for the\ud practitioner, specifically in a time pressured industrial context. Visual information seeking has\ud successfully been applied to reduce the complexity of datasets in healthcare and other fields. Information\ud is presented visually and divided in ‘Overview’, representing the data by its characteristic criteria, and\ud ‘Details’, which are presented on demand. This division reduces information load for the user and eases\ud comparison based on characteristics. This project, first, aims to understand what criteria practitioners use\ud to decide about the suitability of a measure for an in-vehicle evaluation. Secondly, criteria practitioners use\ud to select measures are implemented in a new interface approach based on methods of visual information\ud seeking to support users in the selection and comparison of human-factors related measures for in-vehicle\ud evaluations. Overall, the interface exposes practitioners to new measures, enables them to rapidly compare\ud measures, and obtain information to practically apply them