US 10,890,981 B2
Gesture-based vehicle control
Yifan Chen, Ann Arbor, MI (US); Abhishek Sharma, Novi, MI (US); and Qianyi Wang, Allen Park, MI (US)
Assigned to FORD GLOBAL TECHNOLOGIES, LLC, Dearborn, MI (US)
Appl. No. 16/342,708
Filed by Ford Motor Company, Dearborn, MI (US)
PCT Filed Oct. 24, 2016, PCT No. PCT/US2016/058394
§ 371(c)(1), (2) Date Apr. 17, 2019,
PCT Pub. No. WO2018/080419, PCT Pub. Date May 3, 2018.
Prior Publication US 2019/0294251 A1, Sep. 26, 2019
Int. Cl. G06F 3/01 (2006.01); B60K 37/06 (2006.01); G06F 1/16 (2006.01); B60K 35/00 (2006.01); G06K 9/00 (2006.01)
CPC G06F 3/017 (2013.01) [B60K 35/00 (2013.01); B60K 37/06 (2013.01); G06F 1/163 (2013.01); G06K 9/00375 (2013.01); G06K 9/00536 (2013.01); G06K 9/00832 (2013.01); B60K 2370/146 (2019.05); B60K 2370/1464 (2019.05); B60K 2370/164 (2019.05)] 15 Claims
OG exemplary drawing
 
1. A computer, comprising a processor and a memory, the memory storing instructions executable by the processor to:
receive data from an occupant wearable device;
detect a movement of the occupant wearable device relative to a steering wheel of a vehicle based on a movement classifier at least by detecting (1) a first rotation of the occupant wearable device about a first axis intersecting and defining an angle with a plane defined by a circumference of the steering wheel and (2) a second rotation of the occupant wearable device about a second axis tangent to the circumference of the steering wheel, wherein the movement classifier is created based on ground truth data indicating the movement of the occupant wearable device while an occupant's hand contacts the steering wheel of the vehicle;
cause a first action in the vehicle according to the detected first rotation and a second action according to the detected second rotation;
receive ground truth data synchronized with the data received from the occupant wearable device, the received ground truth data indicating movements of occupant wearable device;
calculate statistical features including at least one of a mean, maximum, minimum and standard deviation of the data received from the occupant wearable device;
determine one or more statistical features that correlate with the movements of the occupant wearable device based on the ground truth data; and
identify movement classifiers based at least on the determined one or more statistical features that correlate with the movements of the occupant wearable device, each movement classifier including at least an association of one of the movements of the occupant wearable device with respective one or more statistical features.