Servo control method and apparatus suitable for the same

1. A servo control method of tracking a predetermined reference trajectory based on an output of a plant, the method comprising:

determining a polynomial indicating a desired reference trajectory and derivatives of the polynomial;

sampling a plant output with respect to a plant input;

generating a reference signal and its derivative signals with reference to a relative time between samples; and

determining a plant input at a sampling time using the generated reference signal and its derivative signals, a delayed plant input and a delayed plant output;

wherein the polynomial is a spline polynomial.

2. The method of claim 1 , wherein the reference trajectory has a ramp pattern, the order of the polynomial is 1, and the derivatives include a non-zero first derivative.

3. The method of claim 1 , wherein the reference trajectory has a square wave pattern, and the derivatives include a non-zero first derivative (velocity) and a non-zero second derivative (acceleration).

4. The method of claim 1 , wherein the reference trajectory has a sinusoidal pattern, and the derivatives are approximated using polynomials of a predetermined time interval.

5. The method of claim 1 , wherein the spline polynomial and its derivatives are determined offline.

6. The method of claim 1 , wherein the determining of the polynomial and its derivatives comprises:

determining a spline polynomial representing the reference trajectory; and

determining derivatives of a spline signal required according to a characteristic of the plant,

wherein the polynomial and derivatives are determined offline.

7. The method of claim 1 , wherein the generating of a reference signal and its derivative signals is performed online.

8. The method of claim 7 , wherein the generating of a reference signal and its derivative signals online comprises:

calculating a relative time between samples;

generating the reference signal with reference to the relative time and the spline polynomial; and

generating the derivative signals of the reference signal using the derivatives.

9. The method of claim 1 , wherein the determining of the plant input is performed using a neural network.

10. The method of claim 9 , wherein the neural network is trained offline by a method of minimizing errors of a plant output.

11. The method of claim 10 , wherein the neural network operates using a feedback control method.

12. The method of claim 10 , wherein the neural network operates using a feed forward control method.

13. The method of claim 10 , wherein the neural network operates using a feedback and feed forward control method.

14. A servo control apparatus for tracking a predetermined reference trajectory based on an output of a plant, the apparatus comprising:

a reference signal generator for generating a reference signal and its derivative signals; and

a neural network for receiving the reference signal and its derivative signals generated by the reference signal generator, a delayed input of the plant and a delayed output of the plant, and for generating an input at a sampling time;

wherein the reference signal generator stores a spline polynomial representing a desirable reference trajectory and its derivatives and generates the reference signal and derivative signals with reference to a sampling time interval.

15. The apparatus of claim 14 , wherein the spline polynomial and its derivatives are determined offline.

16. The apparatus of claim 14 , wherein the reference trajectory has a ramp pattern, the order of the polynomial is 1, and the derivatives include a non-zero first derivative.

17. The apparatus of claim 14 , wherein the reference trajectory has a square wave pattern, and the derivatives include a non-zero first derivative and a non-zero second derivative.

18. The apparatus of claim 14 , wherein the reference trajectory has a sinusoidal pattern, and the derivatives are approximated using polynomials of a predetermined time interval.

19. The apparatus of claim 14 , wherein the neural network is a multi-layer perceptron neural network.

20. The apparatus of claim 19 , wherein the neural network is trained offline to minimize errors of the plant.

21. A computer readable recording medium having recorded thereon a computer readable program for performing a servo control method of tracking a predetermined reference trajectory based on an output of a plant, the method comprising:

determining a polynomial indicating a desired reference trajectory and derivatives of the polynomial;

sampling a plant output with respect to a plant input;

generating a reference signal and its derivative signals with reference to a relative time between samples; and

determining a plant input at a sampling time using the generated reference signal and its derivative signals, a delayed plant input and a delayed plant output;

wherein the polynomial is a spline polynomial.

22. A position control apparatus for controlling a position of a write head writing servo track information, the apparatus comprising:

a reference signal generator for generating a reference signal representing a desirable trajectory of the write head; and

a neural network controlling a position of the write head based on the reference signal and its derivative signals generated by the reference signal generator and location information indicating an actual position of the write head;

wherein the reference signal generator stores a spline polynomial representing a desirable reference trajectory and its derivatives and generates the reference signal and derivative signals with reference to a sampling time interval.