US 11,807,270 B2
State estimator
Armin Stangl, Maisach (DE)
Assigned to Arriver Software AB, San Diego, CA (US)
Appl. No. 16/956,148
Filed by Arriver Software AB, Linköping (SE)
PCT Filed Nov. 22, 2018, PCT No. PCT/EP2018/082261
§ 371(c)(1), (2) Date Jun. 19, 2020,
PCT Pub. No. WO2019/120861, PCT Pub. Date Jun. 27, 2019.
Claims priority of application No. 17208647 (EP), filed on Dec. 19, 2017.
Prior Publication US 2020/0339154 A1, Oct. 29, 2020
Int. Cl. B60W 60/00 (2020.01); B60W 30/09 (2012.01); B60W 30/095 (2012.01); B60W 40/10 (2012.01); G05D 1/00 (2006.01); G05D 1/02 (2020.01); G08G 1/16 (2006.01); G06N 3/045 (2023.01)
CPC B60W 60/0015 (2020.02) [B60W 30/09 (2013.01); B60W 30/0956 (2013.01); B60W 40/10 (2013.01); B60W 60/0027 (2020.02); G05D 1/0088 (2013.01); G05D 1/0221 (2013.01); G06N 3/045 (2023.01); G08G 1/166 (2013.01); G05D 2201/0213 (2013.01)] 15 Claims
OG exemplary drawing
 
1. An apparatus for a motor vehicle driver assistance system for an ego vehicle, the apparatus comprising:
one or more sensors; and
an electronic control unit configured to implement a state estimator configured to use a first state of the ego vehicle to calculate a second state of the ego vehicle by using:
a prediction model to estimate the second state from the first state; and
an update model to refine the estimated second state on the basis of at least one value corresponding to a measurement of the second state, the at least one value being determined from at least one sensor measurement from the one or more sensors;
wherein to calculate the second state from the first state the electronic control unit is configured to use an artificial neural network (“ANN”); and
wherein the update model is an update ANN, and wherein an input vector to the update ANN comprises at least a portion of the estimated second state and the at least one value corresponding to the measurement of the second state, and an output vector of the update ANN is at least a portion of the second state, and wherein the update ANN is configured to correct one or more inaccuracies of the estimated second state produced by the prediction model.