Shirley O’Mara
Pronouns: She/her/hers
Research Mentor(s): Brian Lin
Research Mentor School/College/Department: University of Michigan Transportation Research Institute / Engineering
Program: UROPF
Session: Session 1 (9:00am – 9:50am)
Authors:
Abstract
As automated vehicles (AVs) are introduced into our society, challenges have been brought up. AVs have challenges in mixed traffic scenarios where they are sharing the road with vulnerable road users (VRU) like pedestrians and bicyclists. We need to ensure that everyone will be safe on the roads with AVs. It is becoming apparent that we need to do testing on self-driving vehicles to make sure they detect and avoid collisions with VRU. Our testing was conducted at The University of Michigan’s Mcity test facility. At Mcity we set up scenarios where we had a car with the VRU detection technology and a robot proxy with an adult, child and bicyclist. We tested the difference in detecting the VRU in motion versus being static under more than 30 types of scenarios. We also tested the sensor’s ability to detect the VRU in day time and night time. We are looking at if the cars detection of the distance and missing rate of the VRU differed under different test scenarios. The data is processed to show the pathways of the VRU along with the vehicle and we can analyze the motion and compare it to the data the sensors picked up. We are trying to ensure the safety of VRU with AVs and guarantee that the AVs sensors will be able to detect VRU and prevent collisions. If we can advance the sensors on AVs we can hopefully advance the use of AVs and get more AVs out on the road.
