Driving assistance device and method, and storage medium in which program is stored
A driving assistance device includes a hardware processor and a memory, in which the hardware processor is configured to execute: acquiring image data captured in a vehicle and sensor data relating to a driving operation; converting an image in each frame included in the image data into an image feature amount indicated by the image; combining the image feature amount and the acquired sensor data; performing, based on the combined data, a learning process for a driving assistance information learning model used when a target value relating to the driving operation by a driver of the vehicle is generated from the image feature amount included in the combined data; and generating, when new image data captured in the running vehicle is acquired, the target value relating to the driving operation based on the image feature amount converted from the acquired new image data and the driving assistance information learning model.
1. A driving assistance device, comprising:
one or more processors; and
non-transitory computer readable medium comprising instructions executed by the one or more processors to perform to:
acquire image data captured in a running vehicle, and sensor data relating to a driving operation by a driver of the running vehicle;
convert an image in each frame included in the image data into an image feature amount indicated by the image;
combine the image feature amount converted and the sensor data acquired, the combining including column wise aligning the image feature amounts at times with speeds of the running vehicle from the sensor data at those times;
perform, based on combined data resulting from the combining, a learning process for a driving assistance information learning model used when a target value of a steering wheel angle of the vehicle relating to the driving operation by the driver of the running vehicle is generated from the image feature amount included in the combined data;
store the driving assistance information learning model subjected to learning process;
generate, when new image data captured in the running vehicle is acquired, the target value relating to the driving operation by the driver of the running vehicle, as driving assistance information of the running vehicle, based on the image feature amount converted by the image feature amount converter from the acquired new image data and the driving assistance information learning model stored; and
automatically drive the vehicle based on the driving assistance information.
2. The driving assistance device according to claim 1 , wherein the instructions, when executed, cause the one or more processors to further perform to:
perform a learning process for an image feature amount learning model used for conversion from the image in each frame included in the image data captured in the running vehicle into the image feature amount based on the image in each frame included in the image data acquired; and
store the image feature amount learning model subjected to learning process,
convert the image in each frame acquired into the image feature amount based on the image feature amount learning model stored.
3. The driving assistance device according to claim 2 , wherein
the sensor data is data indicative of a running speed of the running vehicle, and
the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount at a predetermined time and the running speed at the predetermined time, and an objective variable indicating that the running speed at a time after the predetermined time corresponds to any of acceleration, constant speed, or deceleration with respect to the running speed at the predetermined time; and
perform, using the data combined, the learning process for the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
4. The driving assistance device according to claim 3 , wherein
the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount extracted between from a time retroactive from the predetermined time and the predetermined time and the running speed at each time; and
learn, using the data combined, the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
5. The driving assistance device according to claim 2 , wherein
the sensor data is data indicative of a running speed of the running vehicle, and the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount at a predetermined time and the running speed at the predetermined time, and an objective variable indicative of the running speed at a time after the predetermined time; and
perform, using the data combined, the learning process for the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
6. The driving assistance device according to claim 5 , wherein
the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount extracted between from a time retroactive from the predetermined time and the predetermined time and the running speed at each time; and
learn, using the data combined, the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
7. The driving assistance device according to claim 2 , wherein the sensor data includes data indicative of acceleration or an engine speed of the running vehicle.
8. The driving assistance device according to claim 1 , wherein
the sensor data is data indicative of a running speed of the running vehicle, and
the instructions, when executed, cause the one or more processors to perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount at a predetermined time and the running speed at the predetermined time, and an objective variable indicating that the running speed at a time after the predetermined time corresponds to any of acceleration, constant speed, or deceleration with respect to the running speed at the predetermined time; and
perform, using the data combined, the learning process for the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
9. The driving assistance device according to claim 8 , wherein
the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount extracted between a time retroactive from the predetermined time and the predetermined time and the running speed at each time; and
learn, using the data combined, the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
10. The driving assistance device according to claim 1 , wherein
the sensor data is data indicative of a running speed of the running vehicle, and the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined, an explanatory variable indicative of each of the image feature amount at a predetermined time and the running speed at the predetermined time, and an objective variable indicative of the running speed at a time after the predetermined time; and
perform, using the data combined by, the learning process for the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
11. The driving assistance device according to claim 10 , wherein
the instructions, when executed, cause the one or more processors to further perform to:
set, among the data combined an explanatory variable indicative of each of the image feature amount extracted between from a time retroactive from the predetermined time and the predetermined time and the running speed at each time; and
learn, using the data combined, the driving assistance information learning model used when the objective variable is predicted from the explanatory variable.
12. The driving assistance device according to claim 1 , wherein the sensor data includes data indicative of acceleration or an engine speed of the running vehicle.
13. A driving assistance method executed by a driving assistance device, the method comprising:
acquiring image data captured in a running vehicle, and sensor data relating to a driving operation by a driver of the running vehicle;
converting an image in each frame included in the acquired image data into an image feature amount indicated by the image;
combining the converted image feature amount and the acquired sensor data, the combining including column wise aligning the image feature amounts at times with speeds of the running vehicle from the sensor data at those times;
performing, using the combined data, a learning process for a driving assistance information learning model used when a target value of a steering wheel angle of the vehicle relating to the driving operation by the driver of the running vehicle is generated from the image feature amount included in the combined data;
generating, when new image data captured in the running vehicle is acquired, the target value relating to the driving operation by the driver of the vehicle, as driving assistance information of the vehicle, based on the image feature amount converted from the acquired new image data and the driving assistance information learning model; and
automatically driving the vehicle based on the driving assistance information.
14. A non-transitory computer readable medium including instructions that, when executed a hardware processor perform the functions of:
acquiring image data captured in a running vehicle, and sensor data relating to a driving operation by a driver of the running vehicle;
converting an image in each frame included in the acquired image data into an image feature amount indicated by the image;
combining the converted image feature amount and the acquired sensor data, the combining including column wise aligning the image feature amounts at times with speeds of the running vehicle from the sensor data at those times;
performing, using the combined data, a learning process for a driving assistance information learning model used when a target value of a steering wheel angle of the vehicle relating to the driving operation by the driver of the running vehicle is generated from the image feature amount included in the combined data;
generating, when new image data captured in the running vehicle is acquired, the target value relating to the driving operation by the driver of the running vehicle, as driving assistance information of the running vehicle, based on the image feature amount converted from the acquired new image data and the driving assistance information learning model; and
automatically driving the vehicle based on the driving assistance information.