US 12,103,543 B2
Comfort driving data collection system, driving control device, method, and program
Asako Fujii, Tokyo (JP); Yusuke Koitabashi, Tokyo (JP); Takuroh Kashima, Tokyo (JP); Yuki Chiba, Tokyo (JP); and Kenji Sobata, Tokyo (JP)
Assigned to NEC CORPORATION, Tokyo (JP)
Appl. No. 17/627,969
Filed by NEC Corporation, Tokyo (JP)
PCT Filed May 22, 2020, PCT No. PCT/JP2020/020333
§ 371(c)(1), (2) Date Jan. 18, 2022,
PCT Pub. No. WO2021/014738, PCT Pub. Date Jan. 28, 2021.
Claims priority of application No. 2019-133484 (JP), filed on Jul. 19, 2019.
Prior Publication US 2022/0274608 A1, Sep. 1, 2022
Int. Cl. B60W 50/00 (2006.01); B60W 40/08 (2012.01); G06N 20/00 (2019.01)
CPC B60W 50/0098 (2013.01) [B60W 40/08 (2013.01); G06N 20/00 (2019.01); B60W 2540/221 (2020.02)] 9 Claims
OG exemplary drawing
 
1. A comfort driving data collection system comprising:
a memory storing instructions; and
one or more processors configured to execute the instructions to:
learn a comfort determination model, by using comfortable activity data and uncomfortable activity data as first training data, taking an objective variable for a comfort value indicating a degree of comfort, and taking an explanatory variable for each a comfort indicator, where
the comfort indicator measures biological information of an individual when activities are performed,
the comfortable activity data associates the biological information measured by the comfort indicator with a teacher label indicating comfort when an activity classified as a comfortable activity is performed, and
the uncomfortable activity data associates the biological information measured by the comfort indicator with a teacher label indicating discomfort when an activity classified as an uncomfortable activity is performed;
generate, for a subject who is riding in a vehicle, individual data including an explanatory variable and driving situations, the explanatory variable used in the comfort determination model and generated based on the comfort indicator measuring the biological information of the subject during riding in the vehicle obtained in the driving situations;
calculate the comfort value for the subject by applying the comfort determination model to the individual data;
generate driving data indicating a comfortable driving situation and driving data indicating an uncomfortable driving situation according to the calculated comfort value;
determine comfort the subject riding in the vehicle based on a ride model learned by using the generated driving data as second training data; and
control automatic driving of the vehicle based on a result of the determination to increase the comfort of the subject riding in in the vehicle.