‘Before our roads will be solely used by fully automated vehicles, a long transition period is to be expected in which fully automated vehicles, partly automated vehicles and manually-driven vehicles have to share our roads.’ It is the nearby future as SWOV, Institute for Road Safety Research in the Netherlands, sees it. It has researched the interaction between cyclists, pedestrians and automated vehicles.
SWOV has looked into the position of pedestrians and cyclists in such a future traffic system. The report provides an overview of current knowledge, theoretically and empirically, about the interaction of pedestrians and cyclists with (partly) automated vehicles. Furthermore, it identifies what we need to know in order to ensure that an automated driving system, particularly during the transition period, does not compromise the safety of pedestrians and cyclists.
So far, it can be concluded that automated vehicle technology has mainly focused on the detection and recognition of pedestrians and cyclists by the vehicle. Even though good progress has been made, many difficulties are yet to be overcome. Technology to reliably predict intentions and behaviour of pedestrians and cyclists, so that the automated vehicle can accurately adjust its behaviour.
It appears very difficult to predict behavioural intentions of pedestrians and cyclists by current technology.
The decision making and behaviour of pedestrians and cyclists in interaction with (partly) automated cars have received very little attention in the research community. Aspects known to determine current interactions, such as formal rules and regulations, informal rules and non-verbal communication, expectations, and behavioural adaptation are likely to play a different role in a system with (partly) automated vehicles and manually-driven vehicles. If decisions and behaviour of pedestrians and cyclists towards (partly) automated vehicles are found to be different from their behaviour towards a vehicle driven by a human driver, the software developers cannot base their algorithms on what is known about current interactions and behaviour patterns.
These and many other questions need to be answered in order to ensure that further developments towards automated driving will not result in a traffic system that is (even) less safe for pedestrians and cyclists than it is presently.
All the details are to be found in this SWOV-report Safe interaction between cyclists, pedestrians and automated vehicles; What do we know and what do we need to know? www.swov.nl/en/