Driving assistance control system of vehicle

1. A driving assistance control system of a vehicle, the driving assistance control system comprising:

one or more processors configured to:

perform vehicle path following control that controls the vehicle to follow a target vehicle path;

acquire a vehicle feature amount of the vehicle related to the target vehicle path;

perform control so that the vehicle feature amount is input into a trained neural network model;

identify a likelihood of reaching a system limit at which the vehicle path following control for the target vehicle path becomes unsuccessful using an output result of the inputting of the vehicle feature amount into the trained neural network model, wherein the trained neural network model is configured to learn a relationship between a plurality of vehicle feature amounts related to the vehicle path following control and the likelihood of reaching the system limit related to the target vehicle path by machine learning in advance;

output an identification result of the likelihood of reaching the system limit corresponding to the input vehicle feature amounts; and

end the vehicle path following control based on the identification result exceeding a predetermined threshold value for a certain period of time.

2. The driving assistance control system according to claim 1 , wherein the one or more processors is further configured to:

calculate an instruction value to transmit to a traveling device of the vehicle for the vehicle to follow the target vehicle path; and

correct the instruction value according to the identification result.

3. The driving assistance control system according to claim 1 , wherein the one or more processors is further configured to:

calculate the target vehicle path based on the vehicle feature amount; and

correct the target vehicle path according to the identification result.

4. The driving assistance control system according to claim 1 , wherein the one or more processors is further configured to: alarm a driver when the identification result exceeds the predetermined threshold value.

5. The driving assistance control system according to claim 1 , wherein the one or more processors is further configured to:

accumulate past data of the vehicle feature amount over time;

create a spare model of the trained neural network model by machine learning a relationship between the past data and the likelihood of reaching the system limit;

compare a correct answer rate of an identification result by the trained neural network model with a correct answer rate of an identification result by the spare model using test data; and

update the trained neural network model with the spare model when the correct answer rate of the spare model is higher than the correct answer rate of the trained neural network model by a certain level or more.

6. The driving assistance control system according to claim 1 , wherein the one or more processors is further configured to output the identification result when a predetermined confirmation condition is established.

7. The driving assistance control system according to claim 1 , wherein the vehicle feature amount includes at least one of: a lateral deviation between the vehicle and the target vehicle path, a steering angle, a vehicle speed, a vehicle yaw rate, a front-rear acceleration, a lateral acceleration, a steering angle instruction value, a vehicle speed instruction value, a vehicle yaw rate instruction value, a front-rear acceleration instruction value, a lateral acceleration instruction value, or a deviation between an instruction value and a sensor value.

8. The driving assistance control system according to claim 7 , wherein a probability of departing from a lane corresponds to the probability of reaching the system limit.

9. The driving assistance control system according to claim 7 , wherein the system limit includes a vehicle limit event, wherein the vehicle limit event includes one or more of the following events occurring: the vehicle departs from a lane, the vehicle separates a certain level or more from a vehicle target path, a vehicle speed separates a certain level or more from a target speed, the vehicle approaches a certain level or more from a white line or a boundary of a travelable area, or a sensor value of a yaw angle separates a certain level or more from a target value.

10. The driving assistance control system according to claim 1 , wherein a probability of departing from a lane corresponds to the probability of reaching the system limit.

11. The driving assistance control system according to claim 1 , wherein the system limit includes a vehicle limit event, wherein the vehicle limit event includes one or more of the following events occurring: the vehicle departs from a lane, the vehicle separates a certain level or more from a vehicle target path, a vehicle speed separates a certain level or more from a target speed, the vehicle approaches a certain level or more from a white line or a boundary of a travelable area, or a sensor value of a yaw angle separates a certain level or more from a target value.

12. A driving assistance control method of a vehicle, the driving assistance control method comprising:

performing vehicle path following control that controls the vehicle to follow a target vehicle path;

acquiring a vehicle feature amount of the vehicle related to the vehicle path following control;

performing control so that the vehicle feature amount is input into a trained neural network model;

identifying a likelihood of reaching a system limit at which the vehicle path following control for the target vehicle path becomes unsuccessful using an output result of the inputting of the vehicle feature amount into the trained neural network model, wherein the trained neural network model is configured to learn a relationship between a plurality of vehicle feature amounts related to the vehicle path following control and the likelihood of reaching the system limit related to the target vehicle path by machine learning in advance;

output an identification result of the likelihood of reaching the system limit corresponding to the input vehicle feature amounts; and

ending the vehicle path following control based on the identification result exceeding a predetermined threshold value for a preset period of time.

13. The driving assistance control method according to claim 12 , wherein the vehicle feature amount includes at least one of: a lateral deviation between the vehicle and the target vehicle path, a steering angle, a vehicle speed, a vehicle yaw rate, a front-rear acceleration, a lateral acceleration, a steering angle instruction value, a vehicle speed instruction value, a vehicle yaw rate instruction value, a front-rear acceleration instruction value, a lateral acceleration instruction value, or a deviation between an instruction value and a sensor value.

14. The driving assistance control method according to claim 13 , wherein a probability of departing from a lane corresponds to the probability of reaching the system limit.

15. The driving assistance control method according to claim 13 , wherein the system limit includes a vehicle limit event, wherein the vehicle limit event includes one or more of the following events occurring: the vehicle departs from a lane, the vehicle separates a certain level or more from a vehicle target path, a vehicle speed separates a certain level or more from a target speed, the vehicle approaches a certain level or more from a white line or a boundary of a travelable area, or a sensor value of a yaw angle separates a certain level or more from a target value.

16. The driving assistance control method according to claim 12 , wherein a probability of departing from a lane corresponds to the probability of reaching the system limit.

17. The driving assistance control method according to claim 12 , wherein the system limit includes a vehicle limit event, wherein the vehicle limit event includes one or more of the following events occurring: the vehicle departs from a lane, the vehicle separates a certain level or more from a vehicle target path, a vehicle speed separates a certain level or more from a target speed, the vehicle approaches a certain level or more from a white line or a boundary of a travelable area, or a sensor value of a yaw angle separates a certain level or more from a target value.