US 12,296,837 B2
Vehicle judder diagnostic method using artificial intelligence and mobile-based GDS
Jae-Min Jin, Seoul (KR); In-Soo Jung, Goyang-si (KR); and Joon-Hyuk Chang, Seoul (KR)
Assigned to HYUNDAI MOTOR COMPANY, Seoul (KR); KIA CORPORATION, Seoul (KR); and IUCF-HYU (INDUSTRY-UNIVERSITY COOPERATION FOUNDATION HANYANG UNIVERSITY), Seoul (KR)
Filed by HYUNDAI MOTOR COMPANY, Seoul (KR); KIA CORPORATION, Seoul (KR); and IUCF-HYU (INDUSTRY-UNIVERSITY COOPERATION FOUNDATION HANYANG UNIVERSITY), Seoul (KR)
Filed on Nov. 14, 2022, as Appl. No. 17/986,204.
Claims priority of application No. 10-2022-0019998 (KR), filed on Feb. 16, 2022.
Prior Publication US 2023/0256982 A1, Aug. 17, 2023
Int. Cl. B60W 50/04 (2006.01); B60W 40/105 (2012.01); G06N 3/04 (2023.01)
CPC B60W 50/04 (2013.01) [B60W 40/105 (2013.01); G06N 3/04 (2013.01); B60W 2050/041 (2013.01); B60W 2510/0241 (2013.01); B60W 2510/06 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A vehicle judder diagnostic method using artificial intelligence, the vehicle judder diagnostic method comprising:
sampling data that samples input data by using a mobile-based global diagnostic system (GDS) in a test vehicle whose judder occurrence is measured by a judder occurrence determination device and extracts artificial intelligence use data;
building a DNN model that develops the artificial intelligence use data as a judder determination artificial intelligence model through the trial and error of a deep neural network (DNN); and
performing artificial intelligence diagnosis that mounts the judder determination artificial intelligence model in the mobile-based GDS and confirms whether the judder occurs in a judder diagnostic mode of the test vehicle,
wherein the sampling of the data comprises:
preparing test data that builds the test vehicle;
performing GDS measurement that operates the test vehicle in a judder diagnostic mode and performs the sampling by the mobile-based GDS while detecting an accelerometer measurement signal of the test vehicle by the judder occurrence determination device;
extracting data labelling by labelling the input data;
refining the data labelling that removes influence of the size of the data labelling and influence of the sampling period; and
extracting the artificial intelligence use data by classifying the data labelling into learning data and validation verification data, and
wherein the preparing of the test data comprises:
selecting a double clutch transmission (DCT) and a double clutch as a judder test device;
selecting the DCT and the double clutch as the test vehicle; and
defining a judder evaluation for the test vehicle.