[足式机器人]Part3 机构运动学与动力学分析与建模 Ch00-2(4) 质量刚体的在坐标系下运动
本文仅供学习使用,总结很多本现有讲述运动学或动力学书籍后的总结,从矢量的角度进行分析,方法比较传统,但更易理解,并且现有的看似抽象方法,两者本质上并无不同。
2024年底本人学位论文发表后方可摘抄
若有帮助请引用
本文参考:
黎 旭,陈 强 洪,甄 文 强 等.惯 性 张 量 平 移 和 旋 转 复 合 变 换 的 一 般 形 式 及 其 应 用[J].工 程 数 学 学 报,2022,39(06):1005-1011.
食用方法
质量点的动量与角动量
刚体的动量与角动量——力与力矩的关系
惯性矩阵的表达与推导——在刚体运动过程中的作用
惯性矩阵在不同坐标系下的表达
务必自己推导全部公式,并理解每个符号的含义
机构运动学与动力学分析与建模 Ch00-2质量刚体的在坐标系下运动Part4
对 H ? Σ M / O F \vec{H}_{\Sigma _{\mathrm{M}}/\mathrm{O}}^{F} HΣM?/OF?进一步处理可得: H ? Σ M / O F = ∑ i N m P i ? R ? O P i F × ( ω ? F × R ? O P i F ) = ∑ i N m P i ? R ? O P i F × ( ? R ? O P i F × ω ? F ) = ∑ i N m P i ? R ? ~ O P i F ( ? R ? ~ O P i F ) ω ? F \vec{H}_{\Sigma _{\mathrm{M}}/\mathrm{O}}^{F}=\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \vec{R}_{\mathrm{OP}_{\mathrm{i}}}^{F}\times \left( \vec{\omega}^F\times \vec{R}_{\mathrm{OP}_{\mathrm{i}}}^{F} \right)}=\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \vec{R}_{\mathrm{OP}_{\mathrm{i}}}^{F}\times \left( -\vec{R}_{\mathrm{OP}_{\mathrm{i}}}^{F}\times \vec{\omega}^F \right)}=\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \tilde{\vec{R}}_{\mathrm{OP}_{\mathrm{i}}}^{F}\left( -\tilde{\vec{R}}_{\mathrm{OP}_{\mathrm{i}}}^{F} \right)}\vec{\omega}^F HΣM?/OF?=∑iN?mPi???ROPi?F?×(ωF×ROPi?F?)=∑iN?mPi???ROPi?F?×(?ROPi?F?×ωF)=∑iN?mPi???R~OPi?F?(?R~OPi?F?)ωF。进而得出: ? [ I ] = ∑ i N m P i ? R ? ~ O P i F ( ? R ? ~ O P i F ) \Rightarrow \left[ I \right] =\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \tilde{\vec{R}}_{\mathrm{OP}_{\mathrm{i}}}^{F}\left( -\tilde{\vec{R}}_{\mathrm{OP}_{\mathrm{i}}}^{F} \right)} ?[I]=∑iN?mPi???R~OPi?F?(?R~OPi?F?)
2.2.4 牛顿-欧拉方程 Netwon-Euler equation
刚体动力学中常用:
{
F
?
Σ
M
F
=
m
t
o
t
a
l
?
a
?
G
F
M
?
Σ
M
/
G
F
=
[
I
]
Σ
M
/
G
F
α
?
M
F
+
ω
?
M
F
×
(
[
I
]
Σ
M
/
G
F
?
ω
?
M
F
)
\begin{cases} \vec{F}_{\Sigma _{\mathrm{M}}}^{F}=m_{\mathrm{total}}\cdot \vec{a}_{\mathrm{G}}^{F}\\ \vec{M}_{\Sigma _{\mathrm{M}}/\mathrm{G}}^{F}=\left[ I \right] _{\Sigma _{\mathrm{M}}/\mathrm{G}}^{F}\vec{\alpha}_{\mathrm{M}}^{F}+\vec{\omega}_{\mathrm{M}}^{F}\times \left( \left[ I \right] _{\Sigma _{\mathrm{M}}/\mathrm{G}}^{F}\cdot \vec{\omega}_{\mathrm{M}}^{F} \right)\\ \end{cases}
{FΣM?F?=mtotal??aGF?MΣM?/GF?=[I]ΣM?/GF?αMF?+ωMF?×([I]ΣM?/GF??ωMF?)?
2.3 惯性矩阵的转换 Inertia-Matrix Transformation
对于空间中的运动刚体而言,刚体的惯性矩阵一般会根据运动坐标系
{
M
}
??
\left\{ M \right\} \,\,
{M}的基矢量为基底进行计算,而不会直接考虑运动刚体在固定坐标系
{
F
}
??
\left\{ F \right\} \,\,
{F}下的惯性矩阵。此时运动坐标系
{
M
}
??
\left\{ M \right\} \,\,
{M}下计算得出的惯性矩阵记为:
[
I
]
M
\left[ I \right] ^M
[I]M。若运动坐标系
{
M
}
??
\left\{ M \right\} \,\,
{M}与固定坐标系
{
F
}
??
\left\{ F \right\} \,\,
{F}的基矢量满足:
[
i
?
M
j
?
M
k
?
M
]
=
[
Q
M
F
]
T
[
I
^
J
^
K
^
]
\left[ \begin{array}{c} \vec{i}^M\\ \vec{j}^M\\ \vec{k}^M\\ \end{array} \right] =\left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}}\left[ \begin{array}{c} \hat{I}\\ \hat{J}\\ \hat{K}\\ \end{array} \right]
?iMj?MkM?
?=[QMF?]T
?I^J^K^?
?,其中
[
Q
M
F
]
T
\left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}}
[QMF?]T为转换矩阵Transition Matrix
,为正交矩阵Orthogonal Matrix
(满足
[
Q
M
F
]
T
=
[
Q
M
F
]
?
1
=
[
Q
F
M
]
\left[ Q_{\mathrm{M}}^{F} \right] ^T=\left[ Q_{\mathrm{M}}^{F} \right] ^{-1}=\left[ Q_{\mathrm{F}}^{M} \right]
[QMF?]T=[QMF?]?1=[QFM?]),
[
Q
M
F
]
\left[ Q_{\mathrm{M}}^{F} \right]
[QMF?]又称旋转矩阵Rotation~Matrix
(一个向量乘以一个正交阵,相当于对这个向量进行旋转)。也揭示了该矩阵的两个作用:基底转换(转换矩阵
[
Q
M
F
]
T
\left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}}
[QMF?]T)与向量旋转(旋转矩阵
[
Q
M
F
]
\left[ Q_{\mathrm{M}}^{F} \right]
[QMF?]),则考虑最开始的图有:
R
?
P
i
F
=
R
?
M
F
+
[
Q
M
F
]
R
?
P
i
M
\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{F}=\vec{R}_{\mathrm{M}}^{F}+\left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}
RPi?F?=RMF?+[QMF?]RPi?M?
进而分析惯性矩阵,若
O
O
O 点与固定坐标系原点
F
F
F 重合,则有:
[
I
]
Σ
M
F
=
∑
i
N
m
P
i
?
[
(
R
?
P
i
F
)
T
R
?
P
i
F
?
E
?
R
?
P
i
F
(
R
?
P
i
F
)
T
]
=
∑
i
N
m
P
i
?
[
(
R
?
M
F
+
[
Q
M
F
]
R
?
P
i
M
)
T
(
R
?
M
F
+
[
Q
M
F
]
R
?
P
i
M
)
?
E
?
(
R
?
M
F
+
[
Q
M
F
]
R
?
P
i
M
)
(
R
?
M
F
+
[
Q
M
F
]
R
?
P
i
M
)
T
]
=
{
m
t
o
t
a
l
?
[
(
R
?
M
F
)
T
R
?
M
F
?
E
?
R
?
M
F
(
R
?
M
F
)
T
]
?
[
I
1
]
Σ
M
F
+
[
Q
M
F
]
(
∑
i
N
m
P
i
?
[
(
R
?
P
i
M
)
T
R
?
P
i
M
?
E
?
R
?
P
i
M
(
R
?
P
i
M
)
T
]
)
[
Q
M
F
]
T
+
?
[
I
2
]
Σ
M
F
m
t
o
t
a
l
?
[
(
R
?
M
F
)
T
(
[
Q
M
F
]
R
?
C
o
M
M
)
?
E
?
R
?
M
F
(
[
Q
M
F
]
R
?
C
o
M
M
)
T
]
?
[
I
3
]
Σ
M
F
+
m
t
o
t
a
l
?
[
(
[
Q
M
F
]
R
?
C
o
M
M
)
T
R
?
M
F
?
E
?
(
[
Q
M
F
]
R
?
C
o
M
M
)
(
R
?
M
F
)
T
]
?
[
I
4
]
Σ
M
F
=
[
I
1
]
Σ
M
F
+
[
I
2
]
Σ
M
F
+
[
I
3
]
Σ
M
F
+
[
I
4
]
Σ
M
F
\begin{split} \left[ I \right] _{\Sigma _{\mathrm{M}}}^{F}&=\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{F} \right) ^T\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{F}\cdot E-\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{F}\left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{F} \right) ^T \right]} \\ &=\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( \vec{R}_{\mathrm{M}}^{F}+\left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}}\left( \vec{R}_{\mathrm{M}}^{F}+\left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) \cdot E-\left( \vec{R}_{\mathrm{M}}^{F}+\left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) \left( \vec{R}_{\mathrm{M}}^{F}+\left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}} \right]} \\ &=\left\{ \begin{array}{c} \begin{array}{c} \underbrace{m_{\mathrm{total}}\cdot \left[ \left( \vec{R}_{\mathrm{M}}^{F} \right) ^{\mathrm{T}}\vec{R}_{\mathrm{M}}^{F}\cdot E-\vec{R}_{\mathrm{M}}^{F}\left( \vec{R}_{\mathrm{M}}^{F} \right) ^{\mathrm{T}} \right] }\\ \left[ I_1 \right] _{\Sigma _{\mathrm{M}}}^{F}\\ \end{array}+\\ \begin{array}{c} \underbrace{\left[ Q_{\mathrm{M}}^{F} \right] \left( \sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}}\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}\cdot E-\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}\left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}} \right]} \right) \left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}}+}\\ \left[ I_2 \right] _{\Sigma _{\mathrm{M}}}^{F}\\ \end{array}\\ \begin{array}{c} \underbrace{m_{\mathrm{total}}\cdot \left[ \left( \vec{R}_{\mathrm{M}}^{F} \right) ^{\mathrm{T}}\left( \left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{CoM}}^{M} \right) \cdot E-\vec{R}_{\mathrm{M}}^{F}\left( \left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{CoM}}^{M} \right) ^{\mathrm{T}} \right] }\\ \left[ I_3 \right] _{\Sigma _{\mathrm{M}}}^{F}\\ \end{array}+\\ \begin{array}{c} \underbrace{m_{\mathrm{total}}\cdot \left[ \left( \left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{CoM}}^{M} \right) ^T\vec{R}_{\mathrm{M}}^{F}\cdot E-\left( \left[ Q_{\mathrm{M}}^{F} \right] \vec{R}_{\mathrm{CoM}}^{M} \right) \left( \vec{R}_{\mathrm{M}}^{F} \right) ^{\mathrm{T}} \right] }\\ \left[ I_4 \right] _{\Sigma _{\mathrm{M}}}^{F}\\ \end{array}\\ \end{array} \right. \\ &=\left[ I_1 \right] _{\Sigma _{\mathrm{M}}}^{F}+\left[ I_2 \right] _{\Sigma _{\mathrm{M}}}^{F}+\left[ I_3 \right] _{\Sigma _{\mathrm{M}}}^{F}+\left[ I_4 \right] _{\Sigma _{\mathrm{M}}}^{F} \end{split}
[I]ΣM?F??=i∑N?mPi???[(RPi?F?)TRPi?F??E?RPi?F?(RPi?F?)T]=i∑N?mPi???[(RMF?+[QMF?]RPi?M?)T(RMF?+[QMF?]RPi?M?)?E?(RMF?+[QMF?]RPi?M?)(RMF?+[QMF?]RPi?M?)T]=?
?
??
mtotal??[(RMF?)TRMF??E?RMF?(RMF?)T]?[I1?]ΣM?F??+
[QMF?](i∑N?mPi???[(RPi?M?)TRPi?M??E?RPi?M?(RPi?M?)T])[QMF?]T+?[I2?]ΣM?F??
mtotal??[(RMF?)T([QMF?]RCoMM?)?E?RMF?([QMF?]RCoMM?)T]?[I3?]ΣM?F??+
mtotal??[([QMF?]RCoMM?)TRMF??E?([QMF?]RCoMM?)(RMF?)T]?[I4?]ΣM?F???=[I1?]ΣM?F?+[I2?]ΣM?F?+[I3?]ΣM?F?+[I4?]ΣM?F??
其中, [ I 2 ] Σ M F = [ Q M F ] ( ∑ i N m P i ? [ ( R ? P i M ) T R ? P i M ? E ? R ? P i M ( R ? P i M ) T ] ) [ Q M F ] T = [ Q M F ] [ I ] Σ M M [ Q M F ] T \left[ I_2 \right] _{\Sigma _{\mathrm{M}}}^{F}=\left[ Q_{\mathrm{M}}^{F} \right] \left( \sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}}\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}\cdot E-\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}\left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}} \right]} \right) \left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}}=\left[ Q_{\mathrm{M}}^{F} \right] \left[ I \right] _{\Sigma _{\mathrm{M}}}^{M}\left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}} [I2?]ΣM?F?=[QMF?](∑iN?mPi???[(RPi?M?)TRPi?M??E?RPi?M?(RPi?M?)T])[QMF?]T=[QMF?][I]ΣM?M?[QMF?]T,对上式进行讨论:
- 纯回转: 当 R ? M F = 0 \vec{R}_{\mathrm{M}}^{F}=0 RMF?=0时,化简为:
[ I ] Σ M F ∣ R ? M F = 0 = [ I 2 ] Σ M F = [ Q M F ] ( ∑ i N m P i ? [ ( R ? P i M ) T R ? P i M ? E ? R ? P i M ( R ? P i M ) T ] ) [ Q M F ] T = [ Q M F ] [ I ] Σ M M [ Q M F ] T \left. \left[ I \right] _{\Sigma _{\mathrm{M}}}^{F} \right|_{\vec{\mathrm{R}}_{\mathrm{M}}^{F}=0}=\left[ I_2 \right] _{\Sigma _{\mathrm{M}}}^{F}=\left[ Q_{\mathrm{M}}^{F} \right] \left( \sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}}\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}\cdot E-\vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M}\left( \vec{R}_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^{\mathrm{T}} \right]} \right) \left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}}=\left[ Q_{\mathrm{M}}^{F} \right] \left[ I \right] _{\Sigma _{\mathrm{M}}}^{M}\left[ Q_{\mathrm{M}}^{F} \right] ^{\mathrm{T}} [I]ΣM?F? ?RMF?=0?=[I2?]ΣM?F?=[QMF?](i∑N?mPi???[(RPi?M?)TRPi?M??E?RPi?M?(RPi?M?)T])[QMF?]T=[QMF?][I]ΣM?M?[QMF?]T- 纯移动: 当 R ? M F ≠ 0 \vec{R}_{\mathrm{M}}^{F}\ne 0 RMF?=0且 [ Q M F ] = E \left[ Q_{\mathrm{M}}^{F} \right] =E [QMF?]=E时,化简为:
[ I ] Σ M F ∣ R ? M F ≠ 0 , [ Q M F ] = E = [ I 1 ] Σ M F + [ I ] Σ M M \left. \left[ I \right] _{\Sigma _{\mathrm{M}}}^{F} \right|_{\vec{\mathrm{R}}_{\mathrm{M}}^{F}\ne 0,\left[ Q_{\mathrm{M}}^{F} \right] =\mathrm{E}}=\left[ I_1 \right] _{\Sigma _{\mathrm{M}}}^{F}+\left[ I \right] _{\Sigma _{\mathrm{M}}}^{M} [I]ΣM?F? ?RMF?=0,[QMF?]=E?=[I1?]ΣM?F?+[I]ΣM?M?
上式也称为惯性矩阵的平行轴定理Parallel Axis Theorem
。- 运动坐标系原点与质心点重合: 当 R ? C o M F = 0 \vec{R}_{\mathrm{CoM}}^{F}=0 RCoMF?=0时,化简为:
[ I ] F ∣ R ? C o M F = 0 = [ I 1 ] + [ I 2 ] \left. \left[ I \right] ^F \right|_{\vec{R}_{\mathrm{CoM}}^{F}=0}=\left[ I_1 \right] +\left[ I_2 \right] [I]F ?RCoMF?=0?=[I1?]+[I2?]
2.4 惯性矩阵的主轴定理} Principal Axis Theorem
进一步观察惯性矩阵:
[
I
]
M
=
[
∑
i
N
m
P
i
?
[
(
y
P
i
M
)
2
+
(
z
P
i
M
)
2
]
?
∑
i
N
m
P
i
?
x
P
i
M
y
P
i
M
?
∑
i
N
m
P
i
?
(
x
P
i
M
z
P
i
M
)
?
∑
i
N
m
P
i
?
(
y
P
i
M
x
P
i
M
)
∑
i
N
m
P
i
?
[
(
x
P
i
M
)
2
+
(
z
P
i
M
)
2
]
?
∑
i
N
m
P
i
?
(
y
P
i
M
z
P
i
M
)
?
∑
i
N
m
P
i
?
(
z
P
i
M
x
P
i
M
)
?
∑
i
N
m
P
i
?
(
z
P
i
M
y
P
i
M
)
∑
i
N
m
P
i
?
[
(
x
P
i
M
)
2
+
(
y
P
i
M
)
2
]
]
\left[ I \right] ^M=\left[ \begin{matrix} \sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( y_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^2+\left( z_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^2 \right]}& -\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot x_{\mathrm{P}_{\mathrm{i}}}^{M}y_{\mathrm{P}_{\mathrm{i}}}^{M}}& -\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left( x_{\mathrm{P}_{\mathrm{i}}}^{M}z_{\mathrm{P}_{\mathrm{i}}}^{M} \right)}\\ -\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left( y_{\mathrm{P}_{\mathrm{i}}}^{M}x_{\mathrm{P}_{\mathrm{i}}}^{M} \right)}& \sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( x_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^2+\left( z_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^2 \right]}& -\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left( y_{\mathrm{P}_{\mathrm{i}}}^{M}z_{\mathrm{P}_{\mathrm{i}}}^{M} \right)}\\ -\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left( z_{\mathrm{P}_{\mathrm{i}}}^{M}x_{\mathrm{P}_{\mathrm{i}}}^{M} \right)}& -\sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left( z_{\mathrm{P}_{\mathrm{i}}}^{M}y_{\mathrm{P}_{\mathrm{i}}}^{M} \right)}& \sum_i^N{m_{\mathrm{P}_{\mathrm{i}}}\cdot \left[ \left( x_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^2+\left( y_{\mathrm{P}_{\mathrm{i}}}^{M} \right) ^2 \right]}\\ \end{matrix} \right]
[I]M=
?∑iN?mPi???[(yPi?M?)2+(zPi?M?)2]?∑iN?mPi???(yPi?M?xPi?M?)?∑iN?mPi???(zPi?M?xPi?M?)??∑iN?mPi???xPi?M?yPi?M?∑iN?mPi???[(xPi?M?)2+(zPi?M?)2]?∑iN?mPi???(zPi?M?yPi?M?)??∑iN?mPi???(xPi?M?zPi?M?)?∑iN?mPi???(yPi?M?zPi?M?)∑iN?mPi???[(xPi?M?)2+(yPi?M?)2]?
?,为对称矩阵Symmetric Matrix
(此时默认
M
M
M 点与
F
F
F 点重合),则一定能够对角化。
等价于找到另一原点与
M
M
M 重合的坐标系
B
B
B ,使得:
[
I
]
B
=
[
I
x
x
B
0
0
0
I
y
y
B
0
0
0
I
z
z
B
]
\left[ I \right] ^B=\left[ \begin{matrix} I_{\mathrm{xx}}^{B}& 0& 0\\ 0& I_{\mathrm{yy}}^{B}& 0\\ 0& 0& I_{\mathrm{zz}}^{B}\\ \end{matrix} \right]
[I]B=
?IxxB?00?0IyyB?0?00IzzB??
?,根据矩阵对角化Matrix Diagonalizing
的原理,结合纯回转推导可得:
[
I
]
M
=
[
Q
B
M
]
[
I
]
B
[
Q
B
M
]
T
\left[ I \right] ^M=\left[ Q_{\mathrm{B}}^{M} \right] \left[ I \right] ^B\left[ Q_{\mathrm{B}}^{M} \right] ^{\mathrm{T}}
[I]M=[QBM?][I]B[QBM?]T
其中:
- [ Q B M ] \left[ Q_{\mathrm{B}}^{M} \right] [QBM?] 满足 [ i ? B j ? B k ? B ] = [ Q B M ] T [ i ? M j ? M k ? M ] \left[ \begin{array}{c} \vec{i}^B\\ \vec{j}^B\\ \vec{k}^B\\ \end{array} \right] =\left[ Q_{\mathrm{B}}^{M} \right] ^{\mathrm{T}}\left[ \begin{array}{c} \vec{i}^M\\ \vec{j}^M\\ \vec{k}^M\\ \end{array} \right] ?iBj?BkB? ?=[QBM?]T ?iMj?MkM? ?;
- ( I x x B , I y y B , I z z B ) \left( I_{\mathrm{xx}}^{B},I_{\mathrm{yy}}^{B},I_{\mathrm{zz}}^{B} \right) (IxxB?,IyyB?,IzzB?) 为矩阵 [ I ] M \left[ I \right] ^M [I]M的
特征值Eigenvalue
;- [ Q B M ] \left[ Q_{\mathrm{B}}^{M} \right] [QBM?] 为对应于特征值矩阵 [ I ] B \left[ I \right] ^B [I]B的
特征基Standard Eigenvalue Basis
(列向量);
本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。 如若内容造成侵权/违法违规/事实不符,请联系我的编程经验分享网邮箱:veading@qq.com进行投诉反馈,一经查实,立即删除!