Micromobility, and e-scooters in particular, are now present in large numbers in many cities, and projections predict strong demand growth in the post-pandemic period, but e-scooter riders have a high risk of accident/injury. We want to represent human behavior in simulations in a way that can be used to assess proposed infrastructure-based safety countermeasures and to quantitatively predict safety impacts of automated driving. This paper presents an analysis of representative e-scooter rider data in Ingolstadt from a shared e-scooter company and proposes applications to modeling behavior of realistic e-scooter riders. Our analysis considers summary statistics including spatial distributions of trip counts, demand by weekday and time-of-day, as well as trip distance and speed distributions. Further the data reveal a substantial proportion of wrong-way riders (WWR) and riders crossing the road before reaching the traffic light, which can pose severe cognitive challenges to drivers because they are often unexpected. We suspect that automated driving and advanced driver assistance systems could have large safety effects especially in situations that human drivers do not anticipate micromobility users.     «

Micromobility, and e-scooters in particular, are now present in large numbers in many cities, and projections predict strong demand growth in the post-pandemic period, but e-scooter riders have a high risk of accident/injury. We want to represent human behavior in simulations in a way that can be used to assess proposed infrastructure-based safety countermeasures and to quantitatively predict safety impacts of automated driving. This paper presents an analysis of representative e-scooter rider d...     »