正交向量
- 向量在标准正交基中的坐标
a = λ 1 e 1 + λ 2 e 2 + . . . + λ r e r \bold a=\lambda_1\bold e_1+\lambda_2\bold e_2+...+\lambda_r\bold e_r a=λ1e1+λ2e2+...+λrer
λ i = [ a , e i ] \lambda_i=[\bold a,\bold e_i] λi=[a,ei] - 把基 a 1 , a 2 , . . . a r \bold a_1,\bold a_2,...\bold a_r a1,a2,...ar标准正交化
-
施密特正交化
b 1 = a 1 , \bold b_1=\bold a_1, b1=a1,
b 2 = a 2 − [ b 1 , a 2 ] [ b 1 , b 1 ] b 1 , \bold b_2=\bold a_2-\dfrac{[\bold b_1,\bold a_2]}{[\bold b_1,\bold b_1]}\bold b_1, b2=a2−[b1,b1][b1,a2]b1,
说明: [ b 1 , a 2 ] [ b 1 , b 1 ] b 1 \dfrac{[\bold b_1,\bold a_2]}{[\bold b_1,\bold b_1]}\bold b_1 [b1,b1][b1,a2]b1是 a 2 \bold a_2 a2在 b 1 \bold b_1 b1方向上的分量。
[ b 1 , a 2 ] [ b 1 , b 1 ] b 1 = [ a 2 , b 1 ∥ b 1 ∥ ] b 1 ∥ b 1 ∥ \begin{aligned} \dfrac{[\bold b_1,\bold a_2]}{[\bold b_1,\bold b_1]}\bold b_1 &=[\bold a_2,\dfrac{\bold b_1}{\|\bold b_1\|}]\dfrac{\bold b_1}{\|\bold b_1\|} \end{aligned} [b1,b1][b1,a2]b1=[a2,∥b1∥b1]∥b1∥b1
. . . . . . . . . . . . ............ ............
b r = a r − [ b 1 , a r ] [ b 1 , b 1 ] b 1 − [ b 2 , a r ] [ b 2 , b 2 ] b 1 − . . . − [ b r − 1 , a r ] [ b r − 1 , b r − 1 ] b r − 1 . \bold b_r=\bold a_r-\dfrac{[\bold b_1,\bold a_r]}{[\bold b_1,\bold b_1]}\bold b_1-\dfrac{[\bold b_2,\bold a_r]}{[\bold b_2,\bold b_2]}\bold b_1-...-\dfrac{[\bold b_{r-1},\bold a_r]}{[\bold b_{r-1},\bold b_{r-1}]}\bold b_{r-1}. br=ar−[b1,b1][b1,ar]b1−[b2,b2][b2,ar]b1−...−[br−1,br−1][br−1,ar]br−1.
正交验证:
[ b 1 , b 2 ] = [ a 1 , a 2 − [ b 1 , a 2 ] [ b 1 , b 1 ] b 1 ] = [ a 1 , a 2 ] − [ a 1 , [ b 1 , a 2 ] [ b 1 , b 1 ] b 1 ] = [ a 1 , a 2 ] − [ b 1 , a 2 ] [ b 1 , b 1 ] [ a 1 , b 1 ] = [ a 1 , a 2 ] − a 1 T b 1 b 1 T a 2 [ b 1 , b 1 ] = [ a 1 , a 2 ] − a 1 T [ b 1 , b 1 ] a 2 [ b 1 , b 1 ] = 0 \begin{aligned} [\bold b_1,\bold b_2] &=[\bold a_1,\bold a_2-\dfrac{[\bold b_1,\bold a_2]}{[\bold b_1,\bold b_1]}\bold b_1]\\ &=[\bold a_1,\bold a_2]-[\bold a_1,\dfrac{[\bold b_1,\bold a_2]}{[\bold b_1,\bold b_1]}\bold b_1]\\ &=[\bold a_1,\bold a_2]-\dfrac{[\bold b_1,\bold a_2]}{[\bold b_1,\bold b_1]}[\bold a_1,\bold b_1]\\ &=[\bold a_1,\bold a_2]-\dfrac{\bold a_1^T\bold b_1\bold b_1^T\bold a_2}{[\bold b_1,\bold b_1]}\\ &=[\bold a_1,\bold a_2]-\dfrac{\bold a_1^T[\bold b_1,\bold b_1]\bold a_2}{[\bold b_1,\bold b_1]}\\ &=0 \end{aligned} [b1,b2]=[a1,a2−[b1,b1][b1,a2]b1]=[a1,a2]−[a1,[b1,b1][b1,a2]b1]=[a1,a2]−[b1,b1][b1,a2][a1,b1]=[a1,a2]−[b1,b1]a1Tb1b1Ta2=[a1,a2]−[b1,b1]a1T[b1,b1]a2=0 -
单位化
e 1 = 1 ∥ b 1 ∥ b 1 , e 2 = 1 ∥ b 2 ∥ b 2 , . . . , e r = 1 ∥ b r ∥ b r \bold e_1=\dfrac{1}{\|\bold b_1\|}\bold b_1,\bold e_2=\dfrac{1}{\|\bold b_2\|}\bold b_2,...,\bold e_r=\dfrac{1}{\|\bold b_r\|}\bold b_r e1=∥b1∥1b1,e2=∥b2∥1b2,...,er=∥br∥1br.
正交矩阵
- 定义:如果
n
n
n阶矩阵
A
\bold A
A满足
A T A = E ( A − 1 = A T ) \bold A^T\bold A=\bold E(\bold A^{-1}=\bold A^T) ATA=E(A−1=AT)
那么称 A \bold A A为正交矩阵。 - 方阵 A \bold A A为正交矩阵的充分必要条件是 A \bold A A的列(行)向量都是单位向量,且两两正交。
- 性质:
- 若 A \bold A A为正交矩阵,则 A − 1 = A T \bold A^{-1}=\bold A^T A−1=AT也是正交矩阵,且 ∣ A ∣ = 1 或 − 1 |\bold A|=1或-1 ∣A∣=1或−1;
- 如 A \bold A A和 B \bold B B都是正交矩阵,则 A B \bold A\bold B AB也是正交矩阵。
正交变换
- 定义:若 P \bold P P为正交矩阵,则线性变换 y = P x \bold y=\bold P\bold x y=Px称为正交变换。
- 优点: ∥ y ∥ = ∥ x ∥ \|\bold y\|=\|\bold x\| ∥y∥=∥x∥,即正交变换线段长度不变。