With the arrival of Connected and Automated Vehicles (CAVs) concepts for traffic control systems emerge that let CAVs weave over the crossing area of an intersection. However, in the current literature, the safety of these vehicles is generally an assumption rather than a certain
...
With the arrival of Connected and Automated Vehicles (CAVs) concepts for traffic control systems emerge that let CAVs weave over the crossing area of an intersection. However, in the current literature, the safety of these vehicles is generally an assumption rather than a certainty. Moreover, the time period where CAVs and Human-Driven Vehicles (HDVs) share the road (i.e. hybrid traffic period) is often not considered by the design of these concepts.
This paper describes technical and non-technical factors that influence the actual and perceived safety of passengers of HDVs and CAVs during the vehicle interactions that occur in the hybrid traffic period. The impact of the indicated factors on the actual or perceived safety of a vehicle is analyzed and described in terms of required safety margins around a vehicle. The size of the safety margins described depend on the positioning inaccuracy of the vehicle, its trajectory tracking inaccuracy, a time synchronisation inaccuracy, and on the preferred time headway, preferred driver space, and accepted time gap of the car occupants. Due to a lack of available data that could be used to compare the expected size of the safety margins around the vehicle with, this research presents a methodology to generate own data that could be used for this comparison. A virtual reality experiment was executed among 82 participants, which resulted in a 93\% acceptance rate at a time gap of 1.5 seconds between one vehicle leaving and another vehicle entering their conflict area. It appeared that neither the crossing direction (left, right) nor the vehicle composition (HDV, CAV) of a vehicle interaction scenario influenced the perceived safety margins in front of the vehicle. The experiment found lower perceived safety margins in front of the vehicle than was expected according to the literature study. This may be the result of the sample not completely being comparable to the total drivers population. 2.0 seconds was accepted by 99\% of the participants. The question arises what acceptance rate is desired to determine the minimum time gap that can be used as a perceived safety margin by future traffic control systems. The accepted time gap is expected to decrease as soon as more drivers get experienced with CAVs.