US 12,240,474 B2
System and method for correcting in real-time an estimated position of a road object
Vivek Vijaya Kumar, Shelby Township, MI (US); Bo Yu, Troy, MI (US); Donald K. Grimm, Utica, MI (US); Fan Bai, Ann Arbor, MI (US); Joon Hwang, Pflugerville, TX (US); Carl P. Darukhanavala, Royal Oak, MI (US); Carolyn Heather MacLeod, Thornhill (CA); and Muhammad Rehan, Troy, MI (US)
Assigned to GM GLOBAL TECHNOLOGY OPERATIONS LLC, Detroit, MI (US)
Filed by GM Global Technology Operations LLC, Detroit, MI (US)
Filed on Dec. 5, 2022, as Appl. No. 18/061,612.
Prior Publication US 2024/0182045 A1, Jun. 6, 2024
Int. Cl. B60W 50/02 (2012.01); B60W 50/06 (2006.01); G01S 17/86 (2020.01); G01S 19/48 (2010.01); G06T 7/70 (2017.01)
CPC B60W 50/0205 (2013.01) [B60W 50/06 (2013.01); G01S 17/86 (2020.01); G01S 19/485 (2020.05); G06T 7/70 (2017.01); B60W 2420/408 (2024.01); B60W 2520/00 (2013.01); B60W 2554/802 (2020.02); B60W 2555/20 (2020.02); G06T 2207/30252 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A vehicle system in an autonomously controlled vehicle for correcting in real-time a camera-based estimated position of a road object and facilitating a Global Navigation Satellite System (GNSS), the vehicle system comprising:
a camera attached to the vehicle and collecting image sensor data associated with the road object, and the camera generating an image input signal including the image sensor data;
at least one input device for collecting vehicle sensor data associated with a current vehicle position, a current vehicle speed, and a current vehicle heading, and the at least one input device generating a vehicle input signal; and
a computer attached to the vehicle and comprising at least one processor electronically communicating with the camera and the at least one input device, and the computer further comprises a non-transitory computer readable storage medium storing instructions such that the at least one processor is programmed to:
receive the image input signal from the camera and the vehicle input signal from the at least one input device;
match the image sensor data and the vehicle sensor data to one another based on a common time of collection by the associated camera and the associated at least one input device;
determine an error model based on the image sensor data and the vehicle sensor data that are matched to one another;
determine a predicted position of the road object based on the error model, the current vehicle position, the current vehicle speed, and the current vehicle heading;
determine a current camera-based position of the road object based on the image sensor data;
determine a deviation of the current camera-based position from the predicted position;
update the error model based on the deviation by removing a plurality of outlier intersections and determining a mean to reduce an error in the current camera-based position, such that the GNSS is capable of correcting an error in the camera-based estimated position of the road object; and
control the vehicle autonomously based on the updated error model.