US 12,442,660 B2
Automated road incident detection using vehicle sensor data
Leon Oliver Stenneth, Chicago, IL (US); Catalin Bogdan Capota, Palatine, IL (US); Elena-Roxana Fuiorea, Timis (RO); Alexandru Cristian, Timis (RO); and Roxana-Maria Iliese, Jud. Suceava (RO)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Jul. 3, 2023, as Appl. No. 18/217,905.
Claims priority of application No. 202200698 (RO), filed on Oct. 31, 2022.
Prior Publication US 2024/0142264 A1, May 2, 2024
Int. Cl. G01C 21/00 (2006.01); G08G 1/01 (2006.01); G08G 1/052 (2006.01)
CPC G01C 21/3841 (2020.08) [G01C 21/3811 (2020.08); G08G 1/0112 (2013.01); G08G 1/052 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A method, comprising:
obtaining from sensors on a plurality of vehicles, vehicle sensor data from the plurality of vehicles traveling on a road network;
receiving a road segment from the road network;
matching, using a map matching process, the vehicle sensor data to the road segment;
dividing the road segment into a plurality of subsegments;
determining, using the vehicle sensor data as a vehicle travels on the road network, a travel time of the vehicle on two adjacent subsegments of the plurality of subsegments, wherein the two adjacent subsegments includes an upstream subsegment and a downstream subsegment;
identifying a travel time difference between the two adjacent subsegments in response to determining a change in vehicle speed of the vehicle between the two adjacent subsegments, wherein a reduction in speed of the vehicle is detected in response to the travel time of the upstream subsegment is greater than the travel time of the downstream subsegment;
determining, using the travel time, a flow rate that identifies a number of vehicles on the road segment during a time period;
using the vehicle sensor data, the travel time difference, and the flow rate for detecting a road incident on the road segment in response to a confidence for the vehicle sensor data exceeding a threshold;
providing a notification with a location on the road segment where the road incident is detected and a flow rate of the road segment; and
using an alternate source of vehicle data for detecting the road incident on the road segment in response to the confidence for the vehicle sensor data being below the threshold.