Stress and Strain While Driving

Driver assistance and information systems become more and more important in modern vehicles. One general aim of these systems is to optimize driver’s current workload, because overload as well as underload can increase errors, lead to critical situations and is furthermore experienced as unpleasant for the driver. To prevent these, driver information management systems can coordinate the distribution of information to the driver or driver assistance systems can take over parts of the driving task and thus facilitate driving. The major requirement for both of these systems is access to the current driver workload level. In a first study we developed an approach to estimate current driver workload by measuring stress factors (driving manoeuvres, environmental factors, driver intention) using sensor information available in modern cars. Because the approach was developed and tested in real driving, the estimation of strain is based on subjective ratings measured subsequent to the driving task (offline). The objective of the presented simulator study is on the one hand to validate the approach and on the other hand to show that the estimation is also possible with data measured online while driving. The analyses show that the dependencies between the stress factors and the workload level are reproduced. Furthermore, the comparison between online and offline measurement demonstrates that estimation of workload is possible both online and offline.