US 12,366,453 B2
Systems and methods for determining map matched locations on a client device
Raymond Xu, San Francisco, CA (US); Tony Zhang, San Francisco, CA (US); Karina Goot, San Francisco, CA (US); Burak Bostancioglu, Alameda, CA (US); Jun Wu, Oakland, CA (US); Garrett Deland Wells, San Francisco, CA (US); Yanrong Li, Mountain View, CA (US); Benjamin Kin Hoong Low, New York, NY (US); Kerrick Alexander Staley, Jersey City, NJ (US); and James Kevin Murphy, San Francisco, CA (US)
Assigned to Lyft, Inc., San Francisco, CA (US)
Filed by Lyft, Inc., San Francisco, CA (US)
Filed on Jul. 20, 2022, as Appl. No. 17/869,561.
Prior Publication US 2024/0027198 A1, Jan. 25, 2024
Int. Cl. G01C 21/34 (2006.01); G06V 20/10 (2022.01)
CPC G01C 21/3407 (2013.01) [G06V 20/182 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A method implemented by a client device, the method comprising:
obtaining initial sensor data that is indicative of a location of the client device at a first time;
based on the initial sensor data and a road network graph maintained on the client device, determining a set of particles corresponding to candidate road segments within the road network graph, wherein each particle comprises (i) a respective candidate trajectory of the client device along a respective candidate road segment within the road network graph, (ii) a respective candidate position and velocity of the client device along the respective candidate trajectory, and (iii) a respective probability that the particle accurately reflects the location of the client device within the road network graph at the first time;
identifying a particle from the set of particles with a highest probability;
based on the identified particle from the set of particles with the highest probability, determining the location of the client device within the road network graph at the first time;
obtaining new sensor data that is indicative of a location of the client device at a second time;
after obtaining the new sensor data, creating a first updated set of particles by updating each particle in at least a subset of the set of particles to include:
(i) a respective updated candidate trajectory that comprises an extension of the respective candidate trajectory that was previously included in the particle;
(ii) a respective updated candidate position and velocity of the client device along the respective updated candidate trajectory; and
(iii) a respective updated probability that the particle accurately reflects the location of client device within the road network graph at the second time;
identifying a particle from the first updated set of particles with a highest probability; and
based on the identified particle from the first updated set of particles with highest probability, determining the location of the client device within the road network graph at the second time.