﻿1
00:00:00,510 --> 00:00:01,350
‫Welcome back.

2
00:00:01,860 --> 00:00:08,290
‫So let's start with the derivation then, or to be more precise, a recap of the derivation.

3
00:00:09,150 --> 00:00:12,270
‫So this is the screenshot from the previous course.

4
00:00:12,870 --> 00:00:18,510
‫And then what I'm going to do here, I'm going to replace this delta with a U.

5
00:00:19,110 --> 00:00:22,500
‫And also this delta here with the U.

6
00:00:23,460 --> 00:00:31,520
‫So it's going to be Delta U at Karplus I vector and then here Delta U at K plus I vector transposed.

7
00:00:32,160 --> 00:00:39,510
‫Now in our case, in the other case, the horizon period and therefore end equals four.

8
00:00:40,320 --> 00:00:48,040
‫Then here in green you replace the error variables that are here inside the parentheses.

9
00:00:48,690 --> 00:00:53,590
‫So here again, I'm going to replace this delta with a U.

10
00:00:54,210 --> 00:01:02,060
‫And here as well, this is going to be Delta U at K plus I and here Delta you add K plus I transposed.

11
00:01:02,610 --> 00:01:11,670
‫And so when you took this thing here, this parenthesis which was transposed, then you could write

12
00:01:11,670 --> 00:01:14,470
‫it out like this, like we did it in the previous course.

13
00:01:14,490 --> 00:01:18,410
‫So this is a screenshot from the previous course is the same thing.

14
00:01:18,420 --> 00:01:26,760
‫The only difference now is that inside your reference vectors and augment state vectors, you have different

15
00:01:26,760 --> 00:01:34,140
‫variables there and also the dimensions of the matrices, like Kielder, for example, they are different.

16
00:01:34,710 --> 00:01:38,510
‫And also instead of Delta deltas, you have Delta use here.

17
00:01:39,300 --> 00:01:47,520
‫And so when you have this expression transposed, then what you can do, you can rewrite it like this,

18
00:01:47,820 --> 00:01:51,990
‫where you distribute this transpose sine among both terms.

19
00:01:52,740 --> 00:02:01,470
‫And then here, when you open up this parenthesis with this transpose sine, then according to linear

20
00:02:01,470 --> 00:02:03,900
‫algebra, this augment is state vector.

21
00:02:04,260 --> 00:02:07,230
‫That was the second variable here.

22
00:02:07,230 --> 00:02:13,560
‫It will be the first variable here and then the C tilde, it will be the second.

23
00:02:14,280 --> 00:02:19,920
‫And also both of them will be transposed as well.

24
00:02:20,240 --> 00:02:20,760
‫Right.

25
00:02:21,600 --> 00:02:28,260
‫And then you took this entire first term here in purple and then you rewrote it like this.

26
00:02:28,920 --> 00:02:32,010
‫You simply open up the parentheses and that's it.

27
00:02:32,880 --> 00:02:40,080
‫Since your s matrix here was diagonal, you could rewrite this thing like this, which was exactly the

28
00:02:40,080 --> 00:02:43,380
‫same term, like this one here.

29
00:02:44,190 --> 00:02:46,050
‫And then you could put them together.

30
00:02:46,200 --> 00:02:52,440
‫And then that means that this one half here and one half here would disappear.

31
00:02:53,190 --> 00:03:02,070
‫And now, since you have rewritten this term here that I have encircled with this purple line, since

32
00:03:02,070 --> 00:03:08,790
‫you've been able to rewrite it like this, then you could rewrite the entire cost function in this form

33
00:03:09,210 --> 00:03:12,210
‫where I'm just going to make a small correction instead of Delp.

34
00:03:12,210 --> 00:03:15,390
‫I'm going to put you here and also here instead of Delta.

35
00:03:15,390 --> 00:03:16,640
‫I'm going to put you here.

36
00:03:17,040 --> 00:03:23,670
‫So Delta you at K plus I vector and here the same thing only transposed.

