US 12,146,763 B2
Inertial sensor and computer-implemented method for self-calibration of an inertial sensor
Matthias Kuehnel, Boeblingen (DE); and Wenqing Liu, Weil der Stadt (DE)
Assigned to ROBERT BOSCH GMBH, Stuttgart (DE)
Filed by Robert Bosch GmbH, Stuttgart (DE)
Filed on Sep. 25, 2020, as Appl. No. 17/032,672.
Claims priority of application No. 102019214984.7 (DE), filed on Sep. 30, 2019.
Prior Publication US 2021/0095995 A1, Apr. 1, 2021
Int. Cl. G01C 25/00 (2006.01); G01C 19/5726 (2012.01); G01P 21/00 (2006.01); G05B 13/02 (2006.01); G05B 19/404 (2006.01); G06N 3/02 (2006.01); G06N 3/045 (2023.01)
CPC G01C 25/005 (2013.01) [G01C 19/5726 (2013.01); G01P 21/00 (2013.01); G05B 13/027 (2013.01); G05B 13/029 (2013.01); G05B 19/404 (2013.01); G06N 3/02 (2013.01); G06N 3/045 (2023.01)] 9 Claims
OG exemplary drawing
 
1. A computer-implemented method for self-calibration of an inertial sensor, comprising the following steps:
establishing data that relate to the inertial sensor having multiple axes, wherein the data includes an electrical sensitivity for each axis of the multiple axes;
subdividing the data into training data and test data;
setting a first target accuracy value for a first artificial neural network that includes linear and/or non-linear activation functions;
training the first artificial neural network using the training data;
inputting the test data into the trained first artificial neural network to obtain a first output value of the first artificial neural network;
establishing a first output accuracy value based on a comparison result between the first output value and the test data;
storing weightings and the linear and/or non-linear activation functions of the first artificial neural network in a memory unit of the inertial sensor when the first output accuracy value is greater than the first target accuracy value, or training the first artificial neural network again using the training data when the first output accuracy value is less than the first target accuracy value;
establishing an upper limiting value and a lower limiting value for a second artificial neural network that includes non-linear activation functions, based on a predefined constant and on the first output value of the first artificial neural network;
training the second artificial neural network using the training data;
inputting the test data into the trained second artificial neural network to obtain a second output value of the second artificial neural network;
comparing the second output value of the second artificial neural network with a value range from the upper limiting value to the lower limiting value;
establishing a third output value to the second output value when the second output value is within the value range, or to the first output value when the second output value is not within the value range;
storing weightings and the non-linear activation functions of the second artificial neural network and the predefined constant in the memory unit; and
outputting the accuracy value to calibrate the sensor,
wherein, during the training of the first artificial neural network, the first target accuracy value is set as a setpoint value for comparison with the first output accuracy value as an actual value.