控制的一大类对象为以机器人为代表的运动体控制，相应地，我们会比较关心三维空间中被控对象的运动状态，即位置、速度和姿态，对位置和速度的描述需要首先确定坐标系，即是在什么坐标系下的位置和速度，而姿态则描述坐标系与坐标系之间的关系。本部分的主要参考文献为[1]和[2]。
我们通常采用右手坐标系描述运动量，设三维空间中有两个原点重合的坐标系：
s
\mathrm{s}
s系
{
x
^
s
,
y
^
s
,
z
^
s
}
\{\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}\}
{x^s,y^s,z^s}和
b
\mathrm{b}
b系
{
x
^
b
,
y
^
b
,
z
^
b
}
\{\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}\}
{x^b,y^b,z^b}，
x
^
s
,
y
^
s
,
z
^
s
\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}
x^s,y^s,z^s和
x
^
b
,
y
^
b
,
z
^
b
\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}
x^b,y^b,z^b分别为
s
\mathrm{s}
s系和
b
\mathrm{b}
b系下相互正交的三个单位矢量（默认情形下这里的矢量取列矢量形式），且分别与
s
\mathrm{s}
s系和
b
\mathrm{b}
b系下的三个坐标轴正方向平行，对于空间中的任意一个矢量
r
\boldsymbol{r}
r，可以投影到任意坐标系下进行表示，设其在
s
\mathrm{s}
s系下的三个分量为
r
1
s
,
r
2
s
,
r
3
s
r_{1\mathrm{s}},r_{2\mathrm{s}},r_{3\mathrm{s}}
r1s,r2s,r3s，在
b
\mathrm{b}
b系下的三个分量为
r
1
b
,
r
2
b
,
r
3
b
r_{1\mathrm{b}},r_{2\mathrm{b}},r_{3\mathrm{b}}
r1b,r2b,r3b，则有
r
=
r
1
s
x
^
s
+
r
2
s
y
^
s
+
r
3
s
z
^
s
=
r
1
b
x
^
b
+
r
2
b
y
^
b
+
r
3
b
z
^
b
(1)
\boldsymbol{r}=r_{1\mathrm{s}}\hat{\mathbf{x}}_\mathrm{s}+r_{2\mathrm{s}}\hat{\mathbf{y}}_\mathrm{s}+r_{3\mathrm{s}}\hat{\mathbf{z}}_\mathrm{s}=r_{1\mathrm{b}}\hat{\mathbf{x}}_\mathrm{b}+r_{2\mathrm{b}}\hat{\mathbf{y}}_\mathrm{b}+r_{3\mathrm{b}}\hat{\mathbf{z}}_\mathrm{b}\tag{1}
r=r1sx^s+r2sy^s+r3sz^s=r1bx^b+r2by^b+r3bz^b(1)
首先以
s
\mathrm{s}
s系为基准坐标系，对式(1)后半部分的左右两端依次和
x
^
s
,
y
^
s
,
z
^
s
\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}
x^s,y^s,z^s点积，可得
{
r
1
s
=
r
1
b
x
^
b
⋅
x
^
s
+
r
2
b
y
^
b
⋅
x
^
s
+
r
3
b
z
^
b
⋅
x
^
s
r
2
s
=
r
1
b
x
^
b
⋅
y
^
s
+
r
2
b
y
^
b
⋅
y
^
s
+
r
3
b
z
^
b
⋅
y
^
s
r
3
s
=
r
1
b
x
^
b
⋅
z
^
s
+
r
2
b
y
^
b
⋅
z
^
s
+
r
3
b
z
^
b
⋅
z
^
s
(2)
\left\{ \begin{array}{ll} r_{1\mathrm{s}}=r_{1\mathrm{b}}\hat{\mathbf{x}}_\mathrm{b}\cdot\hat{\mathbf{x}}_\mathrm{s}+r_{2\mathrm{b}}\hat{\mathbf{y}}_\mathrm{b}\cdot\hat{\mathbf{x}}_\mathrm{s}+r_{3\mathrm{b}}\hat{\mathbf{z}}_\mathrm{b}\cdot\hat{\mathbf{x}}_\mathrm{s}\\ r_{2\mathrm{s}}=r_{1\mathrm{b}}\hat{\mathbf{x}}_\mathrm{b}\cdot\hat{\mathbf{y}}_\mathrm{s}+r_{2\mathrm{b}}\hat{\mathbf{y}}_\mathrm{b}\cdot\hat{\mathbf{y}}_\mathrm{s}+r_{3\mathrm{b}}\hat{\mathbf{z}}_\mathrm{b}\cdot\hat{\mathbf{y}}_\mathrm{s}\\ r_{3\mathrm{s}}=r_{1\mathrm{b}}\hat{\mathbf{x}}_\mathrm{b}\cdot\hat{\mathbf{z}}_\mathrm{s}+r_{2\mathrm{b}}\hat{\mathbf{y}}_\mathrm{b}\cdot\hat{\mathbf{z}}_\mathrm{s}+r_{3\mathrm{b}}\hat{\mathbf{z}}_\mathrm{b}\cdot\hat{\mathbf{z}}_\mathrm{s} \end{array} \right.\tag{2}
⎩⎨⎧r1s=r1bx^b⋅x^s+r2by^b⋅x^s+r3bz^b⋅x^sr2s=r1bx^b⋅y^s+r2by^b⋅y^s+r3bz^b⋅y^sr3s=r1bx^b⋅z^s+r2by^b⋅z^s+r3bz^b⋅z^s(2)
令
r
s
=
[
r
1
s
,
r
2
s
,
r
3
s
]
T
\boldsymbol{r}^\mathrm{s}=[r_{1\mathrm{s}},r_{2\mathrm{s}},r_{3\mathrm{s}}]^\mathrm{T}
rs=[r1s,r2s,r3s]T和
r
b
=
[
r
1
b
,
r
2
b
,
r
3
b
]
T
\boldsymbol{r}^\mathrm{b}=[r_{1\mathrm{b}},r_{2\mathrm{b}},r_{3\mathrm{b}}]^\mathrm{T}
rb=[r1b,r2b,r3b]T分别表示矢量
r
\boldsymbol{r}
r在与
s
\mathrm{s}
s系和
b
\mathrm{b}
b系下的投影，式(2)可进一步表示为
r
s
=
[
r
1
s
r
2
s
r
3
s
]
=
[
x
^
b
⋅
x
^
s
y
^
b
⋅
x
^
s
z
^
b
⋅
x
^
s
x
^
b
⋅
y
^
s
y
^
b
⋅
y
^
s
z
^
b
⋅
y
^
s
x
^
b
⋅
z
^
s
y
^
b
⋅
z
^
s
z
^
b
⋅
z
^
s
]
[
r
1
b
r
2
b
r
3
b
]
=
C
b
s
r
b
(3)
\begin{aligned} \boldsymbol{r}^\mathrm{s}&=\begin{bmatrix} r_{1\mathrm{s}}\\ r_{2\mathrm{s}}\\ r_{3\mathrm{s}} \end{bmatrix} =\begin{bmatrix} \hat{\mathbf{x}}_\mathrm{b}\cdot\hat{\mathbf{x}}_\mathrm{s} & \hat{\mathbf{y}}_\mathrm{b}\cdot\hat{\mathbf{x}}_\mathrm{s} & \hat{\mathbf{z}}_\mathrm{b}\cdot\hat{\mathbf{x}}_\mathrm{s}\\ \hat{\mathbf{x}}_\mathrm{b}\cdot\hat{\mathbf{y}}_\mathrm{s} & \hat{\mathbf{y}}_\mathrm{b}\cdot\hat{\mathbf{y}}_\mathrm{s} & \hat{\mathbf{z}}_\mathrm{b}\cdot\hat{\mathbf{y}}_\mathrm{s} \\ \hat{\mathbf{x}}_\mathrm{b}\cdot\hat{\mathbf{z}}_\mathrm{s} & \hat{\mathbf{y}}_\mathrm{b}\cdot\hat{\mathbf{z}}_\mathrm{s} & \hat{\mathbf{z}}_\mathrm{b}\cdot\hat{\mathbf{z}}_\mathrm{s} \end{bmatrix}\begin{bmatrix} r_{1\mathrm{b}}\\ r_{2\mathrm{b}}\\ r_{3\mathrm{b}} \end{bmatrix}\\ &=\boldsymbol{C}_\mathrm{b}^\mathrm{s}\boldsymbol{r}^\mathrm{b} \end{aligned} \tag{3}
rs=⎣⎡r1sr2sr3s⎦⎤=⎣⎡x^b⋅x^sx^b⋅y^sx^b⋅z^sy^b⋅x^sy^b⋅y^sy^b⋅z^sz^b⋅x^sz^b⋅y^sz^b⋅z^s⎦⎤⎣⎡r1br2br3b⎦⎤=Cbsrb(3)
式中， C b n \boldsymbol{C}_\mathrm{b}^\mathrm{n} Cbn为从 b \mathrm{b} b系到 s \mathrm{s} s系的旋转矩阵。
然后以
b
\mathrm{b}
b系为基准坐标系，对式(1)后半部分的左右两端依次和
x
^
b
,
y
^
b
,
z
^
b
\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}
x^b,y^b,z^b点积，可得
{
r
1
b
=
r
1
s
x
^
s
⋅
x
^
b
+
r
2
s
y
^
s
⋅
x
^
b
+
r
3
s
z
^
s
⋅
x
^
b
r
2
b
=
r
1
s
x
^
s
⋅
y
^
b
+
r
2
s
y
^
s
⋅
y
^
b
+
r
3
s
z
^
s
⋅
y
^
b
r
3
b
=
r
1
s
x
^
s
⋅
z
^
b
+
r
2
s
y
^
s
⋅
z
^
b
+
r
3
s
z
^
s
⋅
z
^
b
(4)
\left\{ \begin{array}{ll} r_{1\mathrm{b}}=r_{1\mathrm{s}}\hat{\mathbf{x}}_\mathrm{s}\cdot\hat{\mathbf{x}}_\mathrm{b}+r_{2\mathrm{s}}\hat{\mathbf{y}}_\mathrm{s}\cdot\hat{\mathbf{x}}_\mathrm{b}+r_{3\mathrm{s}}\hat{\mathbf{z}}_\mathrm{s}\cdot\hat{\mathbf{x}}_\mathrm{b}\\ r_{2\mathrm{b}}=r_{1\mathrm{s}}\hat{\mathbf{x}}_\mathrm{s}\cdot\hat{\mathbf{y}}_\mathrm{b}+r_{2\mathrm{s}}\hat{\mathbf{y}}_\mathrm{s}\cdot\hat{\mathbf{y}}_\mathrm{b}+r_{3\mathrm{s}}\hat{\mathbf{z}}_\mathrm{s}\cdot\hat{\mathbf{y}}_\mathrm{b}\\ r_{3\mathrm{b}}=r_{1\mathrm{s}}\hat{\mathbf{x}}_\mathrm{s}\cdot\hat{\mathbf{z}}_\mathrm{b}+r_{2\mathrm{s}}\hat{\mathbf{y}}_\mathrm{s}\cdot\hat{\mathbf{z}}_\mathrm{b}+r_{3\mathrm{s}}\hat{\mathbf{z}}_\mathrm{s}\cdot\hat{\mathbf{z}}_\mathrm{b} \end{array} \right.\tag{4}
⎩⎨⎧r1b=r1sx^s⋅x^b+r2sy^s⋅x^b+r3sz^s⋅x^br2b=r1sx^s⋅y^b+r2sy^s⋅y^b+r3sz^s⋅y^br3b=r1sx^s⋅z^b+r2sy^s⋅z^b+r3sz^s⋅z^b(4)
式(4)可进一步表示为
r
b
=
[
r
1
b
r
2
b
r
3
b
]
=
[
x
^
s
⋅
x
^
b
y
^
s
⋅
x
^
b
z
^
s
⋅
x
^
b
x
^
s
⋅
y
^
b
y
^
s
⋅
y
^
b
z
^
s
⋅
y
^
b
x
^
s
⋅
z
^
b
y
^
s
⋅
z
^
b
z
^
s
⋅
z
^
b
]
[
r
1
s
r
2
s
r
3
s
]
=
(
C
b
s
)
T
r
s
=
C
s
b
r
s
(5)
\begin{aligned} \boldsymbol{r}^\mathrm{b}&=\begin{bmatrix} r_{1\mathrm{b}}\\ r_{2\mathrm{b}}\\ r_{3\mathrm{b}} \end{bmatrix} =\begin{bmatrix} \hat{\mathbf{x}}_\mathrm{s}\cdot\hat{\mathbf{x}}_\mathrm{b} & \hat{\mathbf{y}}_\mathrm{s}\cdot\hat{\mathbf{x}}_\mathrm{b} & \hat{\mathbf{z}}_\mathrm{s}\cdot\hat{\mathbf{x}}_\mathrm{b}\\ \hat{\mathbf{x}}_\mathrm{s}\cdot\hat{\mathbf{y}}_\mathrm{b} & \hat{\mathbf{y}}_\mathrm{s}\cdot\hat{\mathbf{y}}_\mathrm{b} & \hat{\mathbf{z}}_\mathrm{s}\cdot\hat{\mathbf{y}}_\mathrm{b} \\ \hat{\mathbf{x}}_\mathrm{s}\cdot\hat{\mathbf{z}}_\mathrm{b} & \hat{\mathbf{y}}_\mathrm{s}\cdot\hat{\mathbf{z}}_\mathrm{b} & \hat{\mathbf{z}}_\mathrm{s}\cdot\hat{\mathbf{z}}_\mathrm{b} \end{bmatrix}\begin{bmatrix} r_{1\mathrm{s}}\\ r_{2\mathrm{s}}\\ r_{3\mathrm{s}} \end{bmatrix}\\ &=(\boldsymbol{C}_\mathrm{b}^\mathrm{s})^\mathrm{T}\boldsymbol{r}^\mathrm{s}=\boldsymbol{C}_\mathrm{s}^\mathrm{b}\boldsymbol{r}^\mathrm{s} \end{aligned}\tag{5}
rb=⎣⎡r1br2br3b⎦⎤=⎣⎡x^s⋅x^bx^s⋅y^bx^s⋅z^by^s⋅x^by^s⋅y^by^s⋅z^bz^s⋅x^bz^s⋅y^bz^s⋅z^b⎦⎤⎣⎡r1sr2sr3s⎦⎤=(Cbs)Trs=Csbrs(5)
式中， C s b = ( C b s ) T \boldsymbol{C}_\mathrm{s}^\mathrm{b}=(\boldsymbol{C}_\mathrm{b}^\mathrm{s})^\mathrm{T} Csb=(Cbs)T为从 s \mathrm{s} s系到 b \mathrm{b} b系的旋转矩阵。
考虑到
x
^
b
,
y
^
b
,
z
^
b
\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}
x^b,y^b,z^b实际上也是三个矢量，因此它们也可以利用
s
\mathrm{s}
s系
{
x
^
s
,
y
^
s
,
z
^
s
}
\{\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}\}
{x^s,y^s,z^s}进行表示，即有
{
x
^
b
=
l
1
x
^
s
+
m
1
y
^
s
+
n
1
z
^
s
y
^
b
=
l
2
x
^
s
+
m
2
y
^
s
+
n
2
z
^
s
z
^
b
=
l
3
x
^
s
+
m
3
y
^
s
+
n
3
z
^
s
(6)
\left\{ \begin{array}{ll} \hat{\mathbf{x}}_\mathrm{b}=l_1\hat{\mathbf{x}}_\mathrm{s}+m_1\hat{\mathbf{y}}_\mathrm{s}+n_1\hat{\mathbf{z}}_\mathrm{s}\\ \hat{\mathbf{y}}_\mathrm{b}=l_2\hat{\mathbf{x}}_\mathrm{s}+m_2\hat{\mathbf{y}}_\mathrm{s}+n_2\hat{\mathbf{z}}_\mathrm{s}\\ \hat{\mathbf{z}}_\mathrm{b}=l_3\hat{\mathbf{x}}_\mathrm{s}+m_3\hat{\mathbf{y}}_\mathrm{s}+n_3\hat{\mathbf{z}}_\mathrm{s} \end{array} \right.\tag{6}
⎩⎨⎧x^b=l1x^s+m1y^s+n1z^sy^b=l2x^s+m2y^s+n2z^sz^b=l3x^s+m3y^s+n3z^s(6)
则式(3)和(5)可进一步表示为
r
s
=
[
r
1
s
r
2
s
r
3
s
]
=
[
l
1
l
2
l
3
m
1
m
2
m
3
n
1
n
2
n
3
]
[
r
1
b
r
2
b
r
3
b
]
=
[
x
^
b
y
^
b
z
^
b
]
r
b
=
C
b
s
r
b
(7)
\begin{aligned} \boldsymbol{r}^\mathrm{s}&=\begin{bmatrix} r_{1\mathrm{s}}\\ r_{2\mathrm{s}}\\ r_{3\mathrm{s}} \end{bmatrix}=\begin{bmatrix} l_1 & l_2 &l_3\\ m_1 & m_2 &m_3\\ n_1 & n_2 &n_3 \end{bmatrix}\begin{bmatrix} r_{1\mathrm{b}}\\ r_{2\mathrm{b}}\\ r_{3\mathrm{b}} \end{bmatrix}\\ &=[\hat{\mathbf{x}}_\mathrm{b}~\hat{\mathbf{y}}_\mathrm{b}~\hat{\mathbf{z}}_\mathrm{b}]\boldsymbol{r}^\mathrm{b}=\boldsymbol{C}_\mathrm{b}^\mathrm{s}\boldsymbol{r}^\mathrm{b} \end{aligned}\tag{7}
rs=⎣⎡r1sr2sr3s⎦⎤=⎣⎡l1m1n1l2m2n2l3m3n3⎦⎤⎣⎡r1br2br3b⎦⎤=[x^b y^b z^b]rb=Cbsrb(7)
r b = [ x ^ b y ^ b z ^ b ] T r s = [ l 1 m 1 n 1 l 2 m 2 n 2 l 3 m 3 n 3 ] r s (8) \boldsymbol{r}^\mathrm{b}=[\hat{\mathbf{x}}_\mathrm{b}~\hat{\mathbf{y}}_\mathrm{b}~\hat{\mathbf{z}}_\mathrm{b}]^\mathrm{T}\boldsymbol{r}^\mathrm{s}=\begin{bmatrix} l_1 & m_1 &n_1\\ l_2 & m_2 &n_2\\ l_3 & m_3 &n_3 \end{bmatrix}\boldsymbol{r}^\mathrm{s}\tag{8} rb=[x^b y^b z^b]Trs=⎣⎡l1l2l3m1m2m3n1n2n3⎦⎤rs(8)
另一方面，考虑到
x
^
s
,
y
^
s
,
z
^
s
\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}
x^s,y^s,z^s实际上也是三个矢量，因此它们也可以利用
b
\mathrm{b}
b系
{
x
^
b
,
y
^
b
,
z
^
b
}
\{\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}\}
{x^b,y^b,z^b}进行表示，且由式(8)可知
{
x
^
s
=
l
1
x
^
b
+
l
2
y
^
b
+
l
3
z
^
b
y
^
s
=
m
1
x
^
b
+
m
2
y
^
b
+
m
3
z
^
b
z
^
s
=
n
1
x
^
b
+
n
2
y
^
b
+
n
3
z
^
b
(9)
\left\{ \begin{array}{ll} \hat{\mathbf{x}}_\mathrm{s}=l_1\hat{\mathbf{x}}_\mathrm{b}+l_2\hat{\mathbf{y}}_\mathrm{b}+l_3\hat{\mathbf{z}}_\mathrm{b}\\ \hat{\mathbf{y}}_\mathrm{s}=m_1\hat{\mathbf{x}}_\mathrm{b}+m_2\hat{\mathbf{y}}_\mathrm{b}+m_3\hat{\mathbf{z}}_\mathrm{b}\\ \hat{\mathbf{z}}_\mathrm{s}=n_1\hat{\mathbf{x}}_\mathrm{b}+n_2\hat{\mathbf{y}}_\mathrm{b}+n_3\hat{\mathbf{z}}_\mathrm{b} \end{array} \right.\tag{9}
⎩⎨⎧x^s=l1x^b+l2y^b+l3z^by^s=m1x^b+m2y^b+m3z^bz^s=n1x^b+n2y^b+n3z^b(9)
由于
x
^
s
,
y
^
s
,
z
^
s
\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}
x^s,y^s,z^s和
x
^
b
,
y
^
b
,
z
^
b
\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}
x^b,y^b,z^b均为相互正交的三个单位矢量，因此有
l
1
2
+
m
1
2
+
n
1
2
=
1
,
l
1
l
2
+
m
1
m
2
+
n
1
n
2
=
0
l
2
2
+
m
2
2
+
n
2
2
=
1
,
l
1
l
3
+
m
1
m
3
+
n
1
n
3
=
0
l
3
2
+
m
3
2
+
n
3
2
=
1
,
l
2
l
3
+
m
2
m
3
+
n
2
n
3
=
0
l
1
2
+
l
2
2
+
l
3
2
=
1
,
l
1
m
1
+
l
2
m
2
+
l
3
m
3
=
0
m
1
2
+
m
2
2
+
m
3
2
=
1
,
l
1
n
1
+
l
2
n
2
+
l
3
n
3
=
0
n
1
2
+
n
2
2
+
n
3
2
=
1
,
m
1
n
1
+
m
2
n
2
+
m
3
n
3
=
0
(10)
\begin{aligned} l_1^2+m_1^2+n_1^2&=1,& l_1l_2+m_1m_2+n_1n_2&=0\\ l_2^2+m_2^2+n_2^2&=1,& l_1l_3+m_1m_3+n_1n_3&=0\\ l_3^2+m_3^2+n_3^2&=1,& l_2l_3+m_2m_3+n_2n_3&=0\\ l_1^2+l_2^2+l_3^2&=1,& l_1m_1+l_2m_2+l_3m_3&=0\\ m_1^2+m_2^2+m_3^2&=1,& l_1n_1+l_2n_2+l_3n_3&=0\\ n_1^2+n_2^2+n_3^2&=1,& m_1n_1+m_2n_2+m_3n_3&=0 \end{aligned}\tag{10}
l12+m12+n12l22+m22+n22l32+m32+n32l12+l22+l32m12+m22+m32n12+n22+n32=1,=1,=1,=1,=1,=1,l1l2+m1m2+n1n2l1l3+m1m3+n1n3l2l3+m2m3+n2n3l1m1+l2m2+l3m3l1n1+l2n2+l3n3m1n1+m2n2+m3n3=0=0=0=0=0=0(10)
由式(10)可知
C
b
s
(
C
b
s
)
T
=
[
l
1
l
2
l
3
m
1
m
2
m
3
n
1
n
2
n
3
]
[
l
1
m
1
n
1
l
2
m
2
n
2
l
3
m
3
n
3
]
=
[
l
1
2
+
l
2
2
+
l
3
2
l
1
m
1
+
l
2
m
2
+
l
3
m
3
l
1
n
1
+
l
2
n
2
+
l
3
n
3
l
1
m
1
+
l
2
m
2
+
l
3
m
3
m
1
2
+
m
2
2
+
m
3
2
m
1
n
1
+
m
2
n
2
+
m
3
n
3
l
1
n
1
+
l
2
n
2
+
l
3
n
3
m
1
n
1
+
m
2
n
2
+
m
3
n
3
n
1
2
+
n
2
2
+
n
3
2
]
=
I
(11)
\begin{aligned} &\boldsymbol{C}_\mathrm{b}^\mathrm{s}(\boldsymbol{C}_\mathrm{b}^\mathrm{s})^\mathrm{T}\\ =&\begin{bmatrix} l_1 & l_2 &l_3\\ m_1 & m_2 &m_3\\ n_1 & n_2 &n_3 \end{bmatrix} \begin{bmatrix} l_1 & m_1 &n_1\\ l_2 & m_2 &n_2\\ l_3 & m_3 &n_3 \end{bmatrix}\\ =&\begin{bmatrix} l_1^2+l_2^2+l_3^2 & l_1m_1+l_2m_2+l_3m_3 & l_1n_1+l_2n_2+l_3n_3\\ l_1m_1+l_2m_2+l_3m_3 & m_1^2+m_2^2+m_3^2 & m_1n_1+m_2n_2+m_3n_3\\ l_1n_1+l_2n_2+l_3n_3 & m_1n_1+m_2n_2+m_3n_3 & n_1^2+n_2^2+n_3^2 \end{bmatrix}\\ =&I \end{aligned}\tag{11}
===Cbs(Cbs)T⎣⎡l1m1n1l2m2n2l3m3n3⎦⎤⎣⎡l1l2l3m1m2m3n1n2n3⎦⎤⎣⎡l12+l22+l32l1m1+l2m2+l3m3l1n1+l2n2+l3n3l1m1+l2m2+l3m3m12+m22+m32m1n1+m2n2+m3n3l1n1+l2n2+l3n3m1n1+m2n2+m3n3n12+n22+n32⎦⎤I(11)
( C b s ) T C b s = [ l 1 m 1 n 1 l 2 m 2 n 2 l 3 m 3 n 3 ] [ l 1 l 2 l 3 m 1 m 2 m 3 n 1 n 2 n 3 ] = [ l 1 2 + m 1 2 + n 1 2 l 1 l 2 + m 1 m 2 + n 1 n 2 l 1 l 3 + m 1 m 3 + n 1 n 3 l 1 l 2 + m 1 m 2 + n 1 n 2 l 2 2 + m 2 2 + n 2 2 l 2 l 3 + m 2 m 3 + n 2 n 3 l 1 l 3 + m 1 m 3 + n 1 n 3 l 2 l 3 + m 2 m 3 + n 2 n 3 l 3 2 + m 3 2 + n 3 2 ] = I (12) \begin{aligned} &(\boldsymbol{C}_\mathrm{b}^\mathrm{s})^\mathrm{T}\boldsymbol{C}_\mathrm{b}^\mathrm{s}\\ =&\begin{bmatrix} l_1 & m_1 &n_1\\ l_2 & m_2 &n_2\\ l_3 & m_3 &n_3 \end{bmatrix} \begin{bmatrix} l_1 & l_2 &l_3\\ m_1 & m_2 &m_3\\ n_1 & n_2 &n_3 \end{bmatrix}\\ =&\begin{bmatrix} l_1^2+m_1^2+n_1^2 & l_1l_2+m_1m_2+n_1n_2 & l_1l_3+m_1m_3+n_1n_3\\ l_1l_2+m_1m_2+n_1n_2 & l_2^2+m_2^2+n_2^2 & l_2l_3+m_2m_3+n_2n_3\\ l_1l_3+m_1m_3+n_1n_3 & l_2l_3+m_2m_3+n_2n_3 & l_3^2+m_3^2+n_3^2 \end{bmatrix}\\ =&I \end{aligned}\tag{12} ===(Cbs)TCbs⎣⎡l1l2l3m1m2m3n1n2n3⎦⎤⎣⎡l1m1n1l2m2n2l3m3n3⎦⎤⎣⎡l12+m12+n12l1l2+m1m2+n1n2l1l3+m1m3+n1n3l1l2+m1m2+n1n2l22+m22+n22l2l3+m2m3+n2n3l1l3+m1m3+n1n3l2l3+m2m3+n2n3l32+m32+n32⎦⎤I(12)
因此有
(
C
b
s
)
T
=
(
C
b
s
)
−
1
(13)
(\boldsymbol{C}_\mathrm{b}^\mathrm{s})^\mathrm{T}=(\boldsymbol{C}_\mathrm{b}^\mathrm{s})^{-1}\tag{13}
(Cbs)T=(Cbs)−1(13)
即
C
b
s
\boldsymbol{C}_\mathrm{b}^\mathrm{s}
Cbs为正交矩阵。进一步考虑到矩阵行列式性质（矩阵乘积的行列式等于矩阵行列式的乘积；矩阵转置后的行列式不变），可知
d
e
t
(
C
b
s
)
=
±
1
\mathbf{det}(\boldsymbol{C}_\mathrm{b}^\mathrm{s})=\pm1
det(Cbs)=±1，因此有
l
1
(
m
2
n
3
−
m
3
n
2
)
−
l
2
(
m
1
n
3
−
m
3
n
1
)
+
l
3
(
m
1
n
2
−
m
2
n
1
)
=
±
1
(14)
l_1(m_2n_3-m_3n_2)-l_2(m_1n_3-m_3n_1)+l_3(m_1n_2-m_2n_1)=\pm 1\tag{14}
l1(m2n3−m3n2)−l2(m1n3−m3n1)+l3(m1n2−m2n1)=±1(14)
进一步考虑到
x
^
b
,
y
^
b
,
z
^
b
\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}
x^b,y^b,z^b均为相互正交的三个单位矢量，相互之间的矢量叉积满足
x
^
b
×
y
^
b
=
z
^
b
,
y
^
b
×
z
^
b
=
x
^
b
,
z
^
b
×
x
^
b
=
y
^
b
(15)
\hat{\mathbf{x}}_\mathrm{b}\times\hat{\mathbf{y}}_\mathrm{b}=\hat{\mathbf{z}}_\mathrm{b},\hat{\mathbf{y}}_\mathrm{b}\times\hat{\mathbf{z}}_\mathrm{b}=\hat{\mathbf{x}}_\mathrm{b},\hat{\mathbf{z}}_\mathrm{b}\times\hat{\mathbf{x}}_\mathrm{b}=\hat{\mathbf{y}}_\mathrm{b}\tag{15}
x^b×y^b=z^b,y^b×z^b=x^b,z^b×x^b=y^b(15)
结合式(9)和(15)可得
∣
x
^
b
y
^
b
z
^
b
l
1
l
2
l
3
m
1
m
2
m
3
∣
=
n
1
x
^
b
+
n
2
y
^
b
+
n
3
z
^
b
∣
x
^
b
y
^
b
z
^
b
m
1
m
2
m
3
n
1
n
2
n
3
∣
=
l
1
x
^
b
+
l
2
y
^
b
+
l
3
z
^
b
∣
x
^
b
y
^
b
z
^
b
n
1
n
2
n
3
l
1
l
2
l
3
∣
=
m
1
x
^
b
+
m
2
y
^
b
+
m
3
z
^
b
(16)
\begin{aligned} \left\vert \begin{matrix} \hat{\mathbf{x}}_\mathrm{b} & \hat{\mathbf{y}}_\mathrm{b} & \hat{\mathbf{z}}_\mathrm{b} \\ l_1 & l_2 & l_3\\ m_1 & m_2 & m_3 \end{matrix} \right\vert&=n_1\hat{\mathbf{x}}_\mathrm{b}+n_2\hat{\mathbf{y}}_\mathrm{b}+n_3\hat{\mathbf{z}}_\mathrm{b}\\ \left\vert \begin{matrix} \hat{\mathbf{x}}_\mathrm{b} & \hat{\mathbf{y}}_\mathrm{b} & \hat{\mathbf{z}}_\mathrm{b} \\ m_1 & m_2 & m_3\\ n_1 & n_2 & n_3 \end{matrix} \right\vert&=l_1\hat{\mathbf{x}}_\mathrm{b}+l_2\hat{\mathbf{y}}_\mathrm{b}+l_3\hat{\mathbf{z}}_\mathrm{b}\\ \left\vert \begin{matrix} \hat{\mathbf{x}}_\mathrm{b} & \hat{\mathbf{y}}_\mathrm{b} & \hat{\mathbf{z}}_\mathrm{b} \\ n_1 & n_2 & n_3\\ l_1 & l_2 & l_3 \end{matrix} \right\vert&=m_1\hat{\mathbf{x}}_\mathrm{b}+m_2\hat{\mathbf{y}}_\mathrm{b}+m_3\hat{\mathbf{z}}_\mathrm{b} \end{aligned}\tag{16}
∣∣∣∣∣∣x^bl1m1y^bl2m2z^bl3m3∣∣∣∣∣∣∣∣∣∣∣∣x^bm1n1y^bm2n2z^bm3n3∣∣∣∣∣∣∣∣∣∣∣∣x^bn1l1y^bn2l2z^bn3l3∣∣∣∣∣∣=n1x^b+n2y^b+n3z^b=l1x^b+l2y^b+l3z^b=m1x^b+m2y^b+m3z^b(16)
对式(16)进一步整理可得
l
1
=
m
2
n
3
−
m
3
n
2
,
l
2
=
m
3
n
1
−
m
1
n
3
,
l
3
=
m
1
n
2
−
m
2
n
1
m
1
=
n
2
l
3
−
n
3
l
2
,
m
2
=
n
3
l
1
−
n
1
l
3
,
m
3
=
n
1
l
2
−
n
2
l
1
n
1
=
l
2
m
3
−
l
3
m
2
,
n
2
=
l
3
m
1
−
l
1
m
3
,
n
3
=
l
1
m
2
−
l
2
m
1
(17)
\begin{aligned} l_1&=m_2n_3-m_3n_2,&l_2&=m_3n_1-m_1n_3,&l_3&=m_1n_2-m_2n_1\\ m_1&=n_2l_3-n_3l_2,&m_2&=n_3l_1-n_1l_3,&m_3&=n_1l_2-n_2l_1\\ n_1&=l_2m_3-l_3m_2,&n_2&=l_3m_1-l_1m_3,&n_3&=l_1m_2-l_2m_1 \end{aligned}\tag{17}
l1m1n1=m2n3−m3n2,=n2l3−n3l2,=l2m3−l3m2,l2m2n2=m3n1−m1n3,=n3l1−n1l3,=l3m1−l1m3,l3m3n3=m1n2−m2n1=n1l2−n2l1=l1m2−l2m1(17)
结合式(10)与式(17)可得
l
1
(
m
2
n
3
−
m
3
n
2
)
−
l
2
(
m
1
n
3
−
m
3
n
1
)
+
l
3
(
m
1
n
2
−
m
2
n
1
)
=
1
(18)
l_1(m_2n_3-m_3n_2)-l_2(m_1n_3-m_3n_1)+l_3(m_1n_2-m_2n_1)=1\tag{18}
l1(m2n3−m3n2)−l2(m1n3−m3n1)+l3(m1n2−m2n1)=1(18)
即 d e t ( C b s ) = 1 \mathbf{det}(\boldsymbol{C}_\mathrm{b}^\mathrm{s})=1 det(Cbs)=1。
我们以 s \mathrm{s} s系为基准系，进一步考虑旋转矩阵 C s b \boldsymbol{C}_\mathrm{s}^\mathrm{b} Csb，根据式(7)和(8)有 C s b = [ x ^ b y ^ b z ^ b ] T \boldsymbol{C}_\mathrm{s}^\mathrm{b}=[\hat{\mathbf{x}}_\mathrm{b}~\hat{\mathbf{y}}_\mathrm{b}~\hat{\mathbf{z}}_\mathrm{b}]^\mathrm{T} Csb=[x^b y^b z^b]T。设 b 1 \mathrm{b_1} b1系为 s \mathrm{s} s系绕 z ^ s \hat{\mathbf{z}}_\mathrm{s} z^s轴正向旋转 θ \theta θ角，如图1所示。
图1 s 系绕 z ^ s 轴正向旋转 θ 角 \text{图1 $\mathrm{s}$系绕$\hat{\mathbf{z}}_\mathrm{s}$轴正向旋转$\theta$角} 图1 s系绕z^s轴正向旋转θ角
则有
{
x
^
b
1
=
cos
(
θ
)
x
^
s
+
sin
(
θ
)
y
^
s
y
^
b
1
=
−
sin
(
θ
)
x
^
s
+
cos
(
θ
)
y
^
s
z
^
b
1
=
z
^
s
(19)
\left\{ \begin{array}{ll} \hat{\mathbf{x}}_{\mathrm{b}_1}=\cos(\theta)\hat{\mathbf{x}}_\mathrm{s}+\sin(\theta)\hat{\mathbf{y}}_\mathrm{s}\\ \hat{\mathbf{y}}_{\mathrm{b}_1}=-\sin(\theta)\hat{\mathbf{x}}_\mathrm{s}+\cos(\theta)\hat{\mathbf{y}}_\mathrm{s}\\ \hat{\mathbf{z}}_{\mathrm{b}_1}=\hat{\mathbf{z}}_\mathrm{s} \end{array} \right.\tag{19}
⎩⎨⎧x^b1=cos(θ)x^s+sin(θ)y^sy^b1=−sin(θ)x^s+cos(θ)y^sz^b1=z^s(19)
因此
C
s
b
1
=
[
x
^
b
1
y
^
b
1
z
^
b
1
]
T
=
[
cos
(
θ
)
sin
(
θ
)
0
−
sin
(
θ
)
cos
(
θ
)
0
0
0
1
]
(20)
\boldsymbol{C}_\mathrm{s}^{\mathrm{b}_1}=[\hat{\mathbf{x}}_{\mathrm{b}_1}~\hat{\mathbf{y}}_{\mathrm{b}_1}~\hat{\mathbf{z}}_{\mathrm{b}_1}]^\mathrm{T}=\begin{bmatrix} \cos(\theta) & \sin(\theta) & 0\\ -\sin(\theta) & \cos(\theta) & 0\\ 0 & 0 & 1 \end{bmatrix}\tag{20}
Csb1=[x^b1 y^b1 z^b1]T=⎣⎡cos(θ)−sin(θ)0sin(θ)cos(θ)0001⎦⎤(20)
我们将式(20)中的旋转矩阵记为
R
(
z
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{z}},\theta)
R(z^,θ)（为简便起见，后续我们将
x
^
s
,
y
^
s
,
z
^
s
\hat{\mathbf{x}}_\mathrm{s},\hat{\mathbf{y}}_\mathrm{s},\hat{\mathbf{z}}_\mathrm{s}
x^s,y^s,z^s简写为
x
^
,
y
^
,
z
^
\hat{\mathbf{x}},\hat{\mathbf{y}},\hat{\mathbf{z}}
x^,y^,z^，在不至于引起混淆的前提下，根据需要也会将
x
^
,
y
^
,
z
^
\hat{\mathbf{x}},\hat{\mathbf{y}},\hat{\mathbf{z}}
x^,y^,z^视为不同坐标系的轴向的单位矢量），类似可得
R
(
x
^
,
θ
)
=
[
1
0
0
0
cos
(
θ
)
sin
(
θ
)
0
−
sin
(
θ
)
cos
(
θ
)
]
,
R
(
y
^
,
θ
)
=
[
cos
(
θ
)
0
−
sin
(
θ
)
0
1
0
sin
(
θ
)
0
cos
(
θ
)
]
(21)
\boldsymbol{R}(\hat{\mathbf{x}},\theta)=\begin{bmatrix} 1 & 0 & 0\\ 0 & \cos(\theta) & \sin(\theta)\\ 0 & -\sin(\theta) & \cos(\theta)\\ \end{bmatrix},~ \boldsymbol{R}(\hat{\mathbf{y}},\theta)=\begin{bmatrix} \cos(\theta) & 0 & -\sin(\theta)\\ 0 & 1 & 0\\ \sin(\theta) & 0 & \cos(\theta)\\ \end{bmatrix}\tag{21}
R(x^,θ)=⎣⎡1000cos(θ)−sin(θ)0sin(θ)cos(θ)⎦⎤, R(y^,θ)=⎣⎡cos(θ)0sin(θ)010−sin(θ)0cos(θ)⎦⎤(21)
注意从
y
^
\hat{\mathbf{y}}
y^轴正向看
x
^
z
^
\hat{\mathbf{x}}\hat{\mathbf{z}}
x^z^平面时，根据右手坐标系的关系可知
x
^
\hat{\mathbf{x}}
x^轴水平朝左，
z
^
\hat{\mathbf{z}}
z^轴水平朝上，逆时针旋转
θ
\theta
θ角后
x
^
\hat{\mathbf{x}}
x^轴跑到水平面下方了，因此
R
(
y
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{y}},\theta)
R(y^,θ)中
sin
(
θ
)
\sin(\theta)
sin(θ)与
−
sin
(
θ
)
-\sin(\theta)
−sin(θ)的位置正好与
R
(
x
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{x}},\theta)
R(x^,θ)和
R
(
z
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{z}},\theta)
R(z^,θ)对称。对于
R
(
z
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{z}},\theta)
R(z^,θ)，有
R
(
z
^
,
θ
)
x
^
=
cos
(
θ
)
x
^
−
sin
(
θ
)
y
^
=
cos
(
θ
)
x
^
−
sin
(
θ
)
z
^
×
x
^
,
R
(
z
^
,
θ
)
y
^
=
cos
(
θ
)
y
^
+
sin
(
θ
)
x
^
=
cos
(
θ
)
y
^
−
sin
(
θ
)
z
^
×
y
^
R
(
z
^
,
θ
)
z
^
=
z
^
(22)
\begin{aligned} \boldsymbol{R}(\hat{\mathbf{z}},\theta)\hat{\mathbf{x}}&=\cos(\theta)\hat{\mathbf{x}}-\sin(\theta)\hat{\mathbf{y}}=\cos(\theta)\hat{\mathbf{x}}-\sin(\theta)\hat{\mathbf{z}}\times\hat{\mathbf{x}},\\ \boldsymbol{R} (\hat{\mathbf{z}},\theta)\hat{\mathbf{y}}&=\cos(\theta)\hat{\mathbf{y}}+\sin(\theta)\hat{\mathbf{x}}=\cos(\theta)\hat{\mathbf{y}}-\sin(\theta)\hat{\mathbf{z}}\times\hat{\mathbf{y}}\\ \boldsymbol{R} (\hat{\mathbf{z}},\theta)\hat{\mathbf{z}}&=\hat{\mathbf{z}} \end{aligned}\tag{22}
R(z^,θ)x^R(z^,θ)y^R(z^,θ)z^=cos(θ)x^−sin(θ)y^=cos(θ)x^−sin(θ)z^×x^,=cos(θ)y^+sin(θ)x^=cos(θ)y^−sin(θ)z^×y^=z^(22)
对
R
(
x
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{x}},\theta)
R(x^,θ)和
R
(
y
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{y}},\theta)
R(y^,θ)也有类似性质。一般地，设
n
^
\hat{\mathbf{n}}
n^为任意单位矢量，
v
⊥
\mathbf{v}_\perp
v⊥为任意矢量
v
\mathbf{v}
v在与
n
^
\hat{\mathbf{n}}
n^垂直平面上的投影，
v
∥
\mathbf{v}_\parallel
v∥为矢量
v
\mathbf{v}
v在
n
^
\hat{\mathbf{n}}
n^方向上的投影，
R
(
n
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{n}},\theta)
R(n^,θ)表示绕
n
^
\hat{\mathbf{n}}
n^正向旋转
θ
\theta
θ角对应的旋转矩阵，则有
R
(
n
^
,
θ
)
v
=
v
∥
+
cos
(
θ
)
v
⊥
−
sin
(
θ
)
n
^
×
v
⊥
=
v
∥
+
cos
(
θ
)
v
⊥
+
sin
(
θ
)
[
[
n
^
]
]
v
⊥
(23)
\begin{aligned} \boldsymbol{R}(\hat{\mathbf{n}},\theta)\mathbf{v}&=\mathbf{v}_\parallel+\cos(\theta)\mathbf{v}_\perp-\sin(\theta)\hat{\mathbf{n}}\times\mathbf{v}_\perp\\ &=\mathbf{v}_\parallel+\cos(\theta)\mathbf{v}_\perp+\sin(\theta)[[\hat{\mathbf{n}}]]\mathbf{v}_\perp \end{aligned}\tag{23}
R(n^,θ)v=v∥+cos(θ)v⊥−sin(θ)n^×v⊥=v∥+cos(θ)v⊥+sin(θ)[[n^]]v⊥(23)
其中
[
[
n
^
]
]
[[\hat{\mathbf{n}}]]
[[n^]]为与
n
^
\hat{\mathbf{n}}
n^对应的反对称阵，即设
n
^
=
[
n
1
,
n
2
,
n
3
]
T
\hat{\mathbf{n}}=[n_1,n_2,n_3]^\mathrm{T}
n^=[n1,n2,n3]T，则有
[
[
n
^
]
]
=
[
0
n
3
−
n
2
−
n
3
0
n
1
n
2
−
n
1
0
]
(24)
[[\hat{\mathbf{n}}]]=\begin{bmatrix} 0 & n_3 & -n_2\\ -n_3 & 0 & n_1\\ n_2 & -n_1 & 0 \end{bmatrix}\tag{24}
[[n^]]=⎣⎡0−n3n2n30−n1−n2n10⎦⎤(24)
对任意两个矢量 a \mathbf{a} a和 b \mathbf{b} b，由式(24)中的定义可知 a × b = − [ [ a ] ] b \mathbf{a}\times\mathbf{b}=-[[\mathbf{a}]]\mathbf{b} a×b=−[[a]]b，有的文献也定义 − [ [ a ] ] -[[\mathbf{a}]] −[[a]]为与 a \mathbf{a} a对应的反对称阵，这个看个人习惯。
考虑矢量三重叉积公式
a
×
(
b
×
c
)
=
(
a
⋅
c
)
b
−
(
a
⋅
b
)
c
(25)
\mathbf{a}\times(\mathbf{b}\times\mathbf{c})=(\mathbf{a}\cdot\mathbf{c})\mathbf{b}-(\mathbf{a}\cdot\mathbf{b})\mathbf{c}\tag{25}
a×(b×c)=(a⋅c)b−(a⋅b)c(25)
在式(25)中取
a
=
b
=
n
^
\mathbf{a}=\mathbf{b}=\hat{\mathbf{n}}
a=b=n^，
c
=
v
\mathbf{c}=\mathbf{v}
c=v，则有
v
=
(
n
^
⋅
v
)
n
^
−
n
^
×
(
n
^
×
v
)
=
n
^
n
^
T
v
−
[
[
n
^
]
]
2
v
=
v
∥
+
v
⊥
(26)
\begin{aligned} \mathbf{v}&=(\hat{\mathbf{n}}\cdot\mathbf{v})\hat{\mathbf{n}}-\hat{\mathbf{n}}\times(\hat{\mathbf{n}}\times\mathbf{v})\\ &=\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T}\mathbf{v}-[[\hat{\mathbf{n}}]]^2\mathbf{v}\\ &=\mathbf{v}_\parallel+\mathbf{v}_\perp \end{aligned}\tag{26}
v=(n^⋅v)n^−n^×(n^×v)=n^n^Tv−[[n^]]2v=v∥+v⊥(26)
结合式(23)和(26)可知
R
(
n
^
,
θ
)
v
=
n
^
n
^
T
v
−
cos
(
θ
)
[
[
n
^
]
]
2
v
−
sin
(
θ
)
[
[
n
^
]
]
3
v
=
n
^
n
^
T
v
−
cos
(
θ
)
[
[
n
^
]
]
2
v
+
sin
(
θ
)
[
[
n
^
]
]
v
(27)
\begin{aligned} \boldsymbol{R}(\hat{\mathbf{n}},\theta)\mathbf{v}&=\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T}\mathbf{v}-\cos(\theta)[[\hat{\mathbf{n}}]]^2\mathbf{v}-\sin(\theta)[[\hat{\mathbf{n}}]]^3\mathbf{v}\\ &=\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T}\mathbf{v}-\cos(\theta)[[\hat{\mathbf{n}}]]^2\mathbf{v}+\sin(\theta)[[\hat{\mathbf{n}}]]\mathbf{v} \end{aligned}\tag{27}
R(n^,θ)v=n^n^Tv−cos(θ)[[n^]]2v−sin(θ)[[n^]]3v=n^n^Tv−cos(θ)[[n^]]2v+sin(θ)[[n^]]v(27)
式(27)中用到了反对称阵的性质
[
[
n
^
]
]
3
=
−
∥
n
^
∥
2
[
[
n
^
]
]
=
−
[
[
n
^
]
]
[[\hat{\mathbf{n}}]]^3=-\Vert\hat{\mathbf{n}}\Vert^2[[\hat{\mathbf{n}}]]=-[[\hat{\mathbf{n}}]]
[[n^]]3=−∥n^∥2[[n^]]=−[[n^]]。由于
v
\mathbf{v}
v为任意矢量，因此有
R
(
n
^
,
θ
)
=
n
^
n
^
T
−
cos
(
θ
)
[
[
n
^
]
]
2
+
sin
(
θ
)
[
[
n
^
]
]
=
cos
(
θ
)
I
+
(
1
−
cos
(
θ
)
)
n
^
n
^
T
+
sin
(
θ
)
[
[
n
^
]
]
(28)
\begin{aligned} \boldsymbol{R}(\hat{\mathbf{n}},\theta)&=\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T}-\cos(\theta)[[\hat{\mathbf{n}}]]^2+\sin(\theta)[[\hat{\mathbf{n}}]]\\ &=\cos(\theta)I+(1-\cos(\theta))\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T}+\sin(\theta)[[\hat{\mathbf{n}}]] \end{aligned}\tag{28}
R(n^,θ)=n^n^T−cos(θ)[[n^]]2+sin(θ)[[n^]]=cos(θ)I+(1−cos(θ))n^n^T+sin(θ)[[n^]](28)
式(28)中用到了反对称阵的性质
[
[
a
]
]
[
[
b
]
]
=
−
(
a
⋅
b
)
I
+
b
a
T
[[\mathbf{a}]][[\mathbf{b}]]=-(\mathbf{a}\cdot\mathbf{b})I+\mathbf{b}\mathbf{a}^\mathrm{T}
[[a]][[b]]=−(a⋅b)I+baT，取
a
=
b
=
n
^
\mathbf{a}=\mathbf{b}=\hat{\mathbf{n}}
a=b=n^可得
[
[
n
^
]
]
2
=
−
I
+
n
^
n
^
T
[[\hat{\mathbf{n}}]]^2=-I+\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T}
[[n^]]2=−I+n^n^T，式(28)也可进一步写为
R
(
n
^
,
θ
)
=
I
+
sin
(
θ
)
[
[
n
^
]
]
+
(
1
−
cos
(
θ
)
)
[
[
n
^
]
]
2
(29)
\boldsymbol{R}(\hat{\mathbf{n}},\theta)=I+\sin(\theta)[[\hat{\mathbf{n}}]]+ (1-\cos(\theta))[[\hat{\mathbf{n}}]]^2\tag{29}
R(n^,θ)=I+sin(θ)[[n^]]+(1−cos(θ))[[n^]]2(29)
将式(29)写成分量形式，可得
R
(
n
^
,
θ
)
=
[
cos
(
θ
)
+
n
1
2
(
1
−
cos
(
θ
)
)
n
1
n
2
(
1
−
cos
(
θ
)
)
+
n
3
sin
(
θ
)
n
1
n
3
(
1
−
cos
(
θ
)
)
−
n
2
sin
(
θ
)
n
2
n
1
(
1
−
cos
(
θ
)
)
−
n
3
sin
(
θ
)
cos
(
θ
)
+
n
2
2
(
1
−
cos
(
θ
)
)
n
2
n
3
(
1
−
cos
(
θ
)
)
+
n
1
sin
(
θ
)
n
3
n
1
(
1
−
cos
(
θ
)
)
+
n
2
sin
(
θ
)
n
3
n
2
(
1
−
cos
(
θ
)
)
−
n
1
sin
(
θ
)
cos
(
θ
)
+
n
3
2
(
1
−
cos
(
θ
)
)
]
(
30
)
\begin{aligned} &\boldsymbol{R}(\hat{\mathbf{n}},\theta)\\ =&\begin{bmatrix} \cos(\theta)+n_1^2(1-\cos(\theta)) & n_1n_2(1-\cos(\theta))+n_3\sin(\theta) & n_1n_3(1-\cos(\theta))-n_2\sin(\theta)\\ n_2n_1(1-\cos(\theta))-n_3\sin(\theta) & \cos(\theta)+n_2^2(1-\cos(\theta)) & n_2n_3(1-\cos(\theta))+n_1\sin(\theta)\\ n_3n_1(1-\cos(\theta))+n_2\sin(\theta) & n_3n_2(1-\cos(\theta))-n_1\sin(\theta) & \cos(\theta)+n_3^2(1-\cos(\theta)) \end{bmatrix}(30) \end{aligned}
=R(n^,θ)⎣⎡cos(θ)+n12(1−cos(θ))n2n1(1−cos(θ))−n3sin(θ)n3n1(1−cos(θ))+n2sin(θ)n1n2(1−cos(θ))+n3sin(θ)cos(θ)+n22(1−cos(θ))n3n2(1−cos(θ))−n1sin(θ)n1n3(1−cos(θ))−n2sin(θ)n2n3(1−cos(θ))+n1sin(θ)cos(θ)+n32(1−cos(θ))⎦⎤(30)
若已知
R
(
n
^
,
θ
)
\boldsymbol{R}(\hat{\mathbf{n}},\theta)
R(n^,θ)，则根据式(30)可知
cos
θ
=
1
2
(
t
r
R
−
1
)
(31)
\cos{\theta}=\frac{1}{2}(\mathrm{tr}\boldsymbol{R}-1)\tag{31}
cosθ=21(trR−1)(31)
当
sin
(
θ
)
≠
0
\sin(\theta)\neq 0
sin(θ)=0时，有
n
^
=
1
2
sin
(
θ
)
[
R
23
−
R
32
R
31
−
R
13
R
12
−
R
21
]
(32)
\hat{\mathbf{n}}=\frac{1}{2\sin(\theta)}\begin{bmatrix} R_{23}-R_{32}\\ R_{31}-R_{13}\\ R_{12}-R_{21} \end{bmatrix}\tag{32}
n^=2sin(θ)1⎣⎡R23−R32R31−R13R12−R21⎦⎤(32)
当 θ = 0 \theta=0 θ=0时， n ^ \hat{\mathbf{n}} n^没有物理意义；当 θ = π \theta=\pi θ=π时， R ( n ^ , π ) = − I + 2 n ^ n ^ T \boldsymbol{R}(\hat{\mathbf{n}},\pi)=-I+2\hat{\mathbf{n}}\hat{\mathbf{n}}^\mathrm{T} R(n^,π)=−I+2n^n^T，此时 R ( n ^ , π ) + I \boldsymbol{R}(\hat{\mathbf{n}},\pi)+I R(n^,π)+I的三个列均与 n ^ \hat{\mathbf{n}} n^平行，而 n ^ \hat{\mathbf{n}} n^的方向则不影响结果。
至此，我们获得了旋转矩阵的一些初步性质。
参考文献
[1]Battin RH. An introduction to the mathematics and methods of astrodynamics, revised edition. American Institute of Aeronautics and Astronautics; 1999.
[2]Shuster MD. A survey of attitude representations. Navigation. 1993;8(9):439-517.
扫一扫
6402
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
