Disk drive adapting feed-forward compensation using iterative learning control over segments of seek length
View Patent ↗A disk drive is disclosed comprising a head actuated over a disk comprising servo information. A position error signal is generated while executing a first plurality of seek operations to seek the head over a seek length, and first feed-forward compensation values are adapted for a first segment of the seek length using an iterative learning control algorithm based on the position error signal. During a second plurality of seeks, second feed-forward compensation values are adapted for a second segment of the seek length using the iterative learning control algorithm based on the position error signal. The servo control system is operable to execute the second plurality of seek operations using the first feed-forward compensation generated for the first segment of the seek length.
1. A disk drive comprising:
a disk comprising servo information;
a head; and
control circuitry comprising a servo control system operable to actuate the head over the disk in response to the servo information, the servo control system operable to:
generate a position error signal while executing a first plurality of seek operations to seek the head over a seek length, and adapt first feed-forward compensation values for a first segment of the seek length using an iterative learning control algorithm based on the position error signal generated over the first segment during the first plurality of seek operations; and
generate the position error signal while executing a second plurality of seek operations to seek the head over the seek length, and adapt second feed-forward compensation values for a second segment of the seek length using the iterative learning control algorithm based on the position error signal generated over the second segment during the second plurality of seek operations;
wherein the servo control system is operable to execute the second plurality of seek operations using the first feed-forward compensation values generated for the first segment of the seek length without adapting the first feed-forward compensation values using the iterative learning control algorithm based on the position error signal generated over the first segment during the second plurality of seek operations.
2. The disk drive as recited in claim 1 , wherein the servo information comprises a plurality of concentric servo sectors extending across substantially an entire radius of the disk.
3. The disk drive as recited in claim 1 , wherein the servo information comprises a plurality of spiral tracks extending across substantially an entire radius of the disk.
4. The disk drive as recited in claim 1 , wherein the servo control system is operable to adapt the first and second feed-forward compensation values according to:
u ilc — j ( i+ 1, k− 2)= u ilc ( i,k− 2)+ K p e ( i,k ),with i= 0, . . . , I j ,and k=k j , . . . ,k j+1 −1
where:
j represents one of the first and second segments of the seek length;
i represents a seek index for a current seek operation out of one of the first and second plurality I j of seek operations;
k represents an index of the position error signal generated over one of the first and second segments of the seek length;
e(i,k) represents the position error signal;
u ilc — j represents one of the first and second feed-forward compensation values; and
K p represents a gain of the iterative learning control algorithm.
5. The disk drive as recited in claim 1 , wherein the servo control system is operable to adapt the first and second feed-forward compensation values according to:
u ilc — j ( i+ 1, k− 2)= u ilc ( i,k− 2)+ K pj e ( i,k ),with i= 0, . . . , I j ,and k=k j , . . . ,k j+1 −1
where:
j represents one of the first and second segments of the seek length;
i represents a seek index for a current seek operation out of one of the first and second plurality I j of seek operations;
k represents an index of the position error signal generated over one of the first and second segments of the seek length;
e(i,k) represents the position error signal;
u ilc — j represents one of the first and second feed-forward compensation values; and
K pj represents a gain of the iterative learning control algorithm corresponding to one of the first and second segments of the seek length.
6. The disk drive as recited in claim 5 , wherein K pj for the first segment is different than K pj for the second segment.
7. The disk drive as recited in claim 6 , wherein K pj for the first segment is greater than K pj for the second segment.
8. The disk drive as recited in claim 7 , wherein the first segment precedes the second segment.
9. The disk drive as recited in claim 1 , wherein the servo control system is operable to adapt the first feed-forward compensation values over the first plurality of seek operations until the first feed-forward compensation values converge.
10. The disk drive as recited in claim 9 , wherein the first feed-forward compensation values converge when an average position error signal over the first segment falls below a threshold.
11. A method of operating a disk drive comprising a head actuated over a disk comprising servo information, and a servo control system operable to actuate the head over the disk in response to the servo information, the method comprising:
generating a position error signal while executing a first plurality of seek operations to seek the head over a seek length, and adapting first feed-forward compensation values for a first segment of the seek length using an iterative learning control algorithm based on the position error signal generated over the first segment during the first plurality of seek operations; and
generating the position error signal while executing a second plurality of seek operations to seek the head over the seek length, and adapting second feed-forward compensation values for a second segment of the seek length using the iterative learning control algorithm based on the position error signal generated over the second segment during the second plurality of seek operations;
wherein the servo control system is operable to execute the second plurality of seek operations using the first feed-forward compensation values generated for the first segment of the seek length without adapting the first feed-forward compensation values using the iterative learning control algorithm based on the position error signal generated over the first segment during the second plurality of seek operations.
12. The method as recited in claim 11 , wherein the servo information comprises a plurality of concentric servo sectors extending across substantially an entire radius of the disk.
13. The method as recited in claim 11 , wherein the servo information comprises a plurality of spiral tracks extending across substantially an entire radius of the disk.
14. The method as recited in claim 11 , wherein the first and second feed-forward compensation values are adapted according to:
u ilc — j ( i+ 1, k− 2)= u ilc ( i,k− 2)+ K p e ( i,k ),with i= 0, . . . , I j ,and k=k j , . . . ,k j+1 −1
where:
j represents one of the first and second segments of the seek length;
i represents a seek index for a current seek operation out of one of the first and second plurality I j of seek operations;
k represents an index of the position error signal generated over one of the first and second segments of the seek length;
e(i,k) represents the position error signal;
u ilc — j represents one of the first and second feed-forward compensation values; and
K p represents a gain of the iterative learning control algorithm.
15. The method as recited in claim 11 , wherein the first and second feed-forward compensation values are adapted according to:
u ilc — j ( i+ 1, k− 2)= u ilc ( i,k− 2)+ K pj e ( i,k ),with i= 0, . . . , I j ,and k=k j , . . . ,k j+1 −1
where:
j represents one of the first and second segments of the seek length;
i represents a seek index for a current seek operation out of one of the first and second plurality I j of seek operations;
k represents an index of the position error signal generated over one of the first and second segments of the seek length;
e(i,k) represents the position error signal;
u ilc — j represents one of the first and second feed-forward compensation values; and
K pj represents a gain of the iterative learning control algorithm corresponding to one of the first and second segments of the seek length.
16. The method as recited in claim 15 , wherein K pj for the first segment is different than K pj for the second segment.
17. The method as recited in claim 16 , wherein K pj for the first segment is greater than K pj for the second segment.
18. The method as recited in claim 17 , wherein the first segment precedes the second segment.
19. The method as recited in claim 11 , further comprising adapting the first feed-forward compensation values over the first plurality of seek operations until the first feed-forward compensation values converge.
20. The method as recited in claim 19 , wherein the first feed-forward compensation values converge when an average position error signal over the first segment falls below a threshold.