US 11,986,916 B2
High-precision kickback detection for power tools
Chad Thomas Green, Ashburn, VA (US); Ty Batchelor, Decatur, AL (US); and Destin Sandlin, Hartselle, AL (US)
Assigned to LATERN HOLDINGS, LLC, Sterling, VA (US)
Filed by Lantern Holdings, LLC, Sterling, VA (US)
Filed on May 20, 2020, as Appl. No. 16/878,975.
Application 16/878,975 is a continuation of application No. PCT/US2019/018967, filed on Feb. 21, 2019.
Claims priority of provisional application 62/633,305, filed on Feb. 21, 2018.
Prior Publication US 2020/0276680 A1, Sep. 3, 2020
Int. Cl. B23Q 11/00 (2006.01); G06N 20/00 (2019.01); H02K 7/14 (2006.01); H02K 11/21 (2016.01); B23D 59/00 (2006.01); G01P 3/36 (2006.01); G01P 3/44 (2006.01); G01P 15/18 (2013.01); G05B 19/406 (2006.01)
CPC B23Q 11/0092 (2013.01) [G06N 20/00 (2019.01); H02K 7/14 (2013.01); H02K 11/21 (2016.01); B23D 59/001 (2013.01); G01P 3/36 (2013.01); G01P 3/44 (2013.01); G01P 15/18 (2013.01); G05B 19/406 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A multi-dimensional kickback detection system for a power tool comprising:
a three-axes accelerometer sensor;
a three-axes gyroscope sensor;
a three-axes magnetometer measuring the magnetic flux of an electric motor of the power tool; and
a processor comprising a machine learning model pre-trained for the power tool to predict an occurrence of a kickback event of the power tool, the machine learning model using data received from the accelerometer, gyroscope, and magnetometer to predict an occurrence of the kickback event, wherein the processor activates a reactive device of the power tool upon prediction of the kickback event.