Apparatus for controlling gear shift of vehicle and method thereof

1. An apparatus of controlling a gear shifting of a vehicle, the apparatus comprising:

a storage configured to store a Gaussian process (GP) model on which machine learning is completed; and

a controller configured to detect a change amount of an engine torque and a change amount of an engagement-side clutch torque according to the GP model, and control the gear shifting of the vehicle according to the change amount of the engine torque and the change amount of the engagement-side clutch torque.

2. The apparatus of claim 1 , wherein the controller is configured to input, to the GP model, an engine speed, an input shaft speed of a transmission in the vehicle, an engagement-side clutch slip speed, the engine torque, the engagement-side clutch torque, a torsion angle of a driveshaft detected at a previous time point, and the change amount of the engine torque and the change amount of the engagement-side clutch torque detected at a current time point, and obtain the engine torque and the engagement-side clutch torque as an output of the GP model.

3. The apparatus of claim 2 , further including:

a wheel speed sensor configured to detect a wheel speed of the vehicle;

a gear stage sensor configured to detect a current gear stage of the vehicle; and

a clutch sensor configured to detect an engagement-side clutch speed of the vehicle.

4. The apparatus of claim 3 , wherein the controller electrically connected to the clutch sensor is configured to determine the engagement-side clutch slip speed according to a difference between the engine speed and the engagement-side clutch speed detected by the clutch sensor.

5. The apparatus of claim 3 , wherein the controller electrically connected to the wheel speed sensor and the gear stage sensor is configured to determine the input shaft speed of the transmission according to the wheel speed detected by the wheel speed sensor and a gear ratio corresponding to the current gear stage detected by the gear stage sensor.

6. The apparatus of claim 3 , wherein the controller electrically connected to the wheel speed sensor and the clutch sensor is configured to determine the torsion angle of the driveshaft according to the wheel speed detected by the wheel speed sensor and the engagement-side clutch speed detected by the clutch sensor.

7. The apparatus of claim 1 , wherein the controller is configured to obtain a solution of a model predictive control (MPC) optimization problem according to the GP model to detect the change amount of the engine torque and the change amount of the engagement-side clutch torque.

8. The apparatus of claim 1 , wherein the controller is configured to adjust the engine torque and the engagement-side clutch torque in an inertia phase.

9. The apparatus of claim 1 , wherein the GP model includes a mean function which is a zero-mean type, a covariance function which is a squared exponential automatic relevance determination, and uncertainty propagation which is a Taylor approximation.

10. A method of controlling a gear shifting of a vehicle, the method comprising:

storing, by a storage, a Gaussian process (GP) model on which machine learning is completed;

detecting, by a controller, a change amount of an engine torque and a change amount of an engagement-side clutch torque according to the GP model; and

controlling, by the controller, the gear shifting of the vehicle according to the change amount of the engine torque and the change amount of the engagement-side clutch torque.

11. The method of claim 10 , wherein the detecting of the change amount of the engine torque and the change amount of the engagement-side clutch torque includes:

inputting, to the GP model, an engine speed, an input shaft speed of a transmission in the vehicle, an engagement-side clutch slip speed, the engine torque, the engagement-side clutch torque, a torsion angle of a driveshaft detected at a previous time point, and the change amount of the engine torque and the change amount of the engagement-side clutch torque detected at a current time point; and

obtaining the engine torque and the engagement-side clutch torque as an output of the GP model.

12. The method of claim 11 , further including:

detecting, by a wheel speed sensor, a wheel speed of the vehicle;

detecting, by a gear stage sensor, a current gear stage of the vehicle; and

detecting, by a clutch sensor, an engagement-side clutch speed of the vehicle.

13. The method of claim 12 , wherein the inputting to the GP model includes:

determining the engagement-side clutch slip speed according to a difference between the engine speed and the engagement-side clutch speed detected by the clutch sensor.

14. The method of claim 12 , wherein the inputting to the GP model includes:

determining the input shaft speed of the transmission according to the wheel speed detected by the wheel speed sensor and a gear ratio corresponding to the current gear stage detected by the gear stage sensor.

15. The method of claim 12 , wherein the inputting to the GP model includes:

determining the torsion angle of the driveshaft according to the wheel speed detected by the wheel speed sensor and the engagement-side clutch speed detected by the clutch sensor.

16. The method of claim 10 , wherein the controlling of the gear shifting of the vehicle includes:

obtaining a solution of a model predictive control (MPC) optimization problem according to the GP model to detect the change amount of the engine torque and the change amount of the engagement-side clutch torque.

17. The method of claim 10 , wherein the controlling of the gear shifting of the vehicle includes:

adjusting the engine torque and the engagement-side clutch torque in an inertia phase.

18. The method of claim 10 , wherein the GP model includes a mean function which is a zero-mean type, a covariance function which is a squared exponential automatic relevance determination, and uncertainty propagation which is a Taylor approximation.