SLAM从入门到放弃:SLAM十四讲第四章习题(1-4)

以下均为简单笔记,如有错误,请多多指教。

  1. 验证 S O ( 3 ) SO(3) SO(3) S E ( 3 ) SE(3) SE(3) S i m ( 3 ) Sim(3) Sim(3)关于乘法成群。
    证:
    已知 S O ( 3 ) = { R ∈ R 3 × 3 ∣ R R T = I , d e t ( R ) = 1 } SO(3)=\lbrace{R\isin\R^{3\times3}| RR^T=I,det(R)=1}\rbrace SO(3)={RR3×3RRT=I,det(R)=1};取 R 1 ∈ S O ( 3 ) R_1\isin SO(3) R1SO(3) R 2 ∈ S O ( 3 ) R_2\isin SO(3) R2SO(3) R 3 ∈ S O ( 3 ) R_3\isin SO(3) R3SO(3);
    验证封闭性:设 R = R 1 R 2 R=R_1R_2 R=R1R2;不难发现 R R T = ( R 1 R 2 ) ( R 1 R 2 ) T = R 1 R 2 R 2 T R 1 T = I RR^T=(R_1R_2)(R_1R_2)^T=R_1R_2R_2^TR_1^T=I RRT=(R1R2)(R1R2)T=R1R2R2TR1T=I;并且 d e t ( R ) = d e t ( R 1 ) d e t ( R 2 ) = 1 det(R)=det(R_1)det(R_2)=1 det(R)=det(R1)det(R2)=1;故 R ∈ S O ( 3 ) R\isin SO(3) RSO(3)
    验证结合律:根据矩阵乘法结合律,不难得到 ( R 1 R 2 ) R 3 = R 1 ( R 2 R 3 ) (R_1R_2)R_3=R_1(R_2R_3) (R1R2)R3=R1(R2R3)
    幺元:根据单位矩阵的特性,可以发现 I R = R I = R IR=RI=R IR=RI=R
    逆:由于单位旋转矩阵具有正定性,即 R R T = I RR^T=I RRT=I

    已知 S E ( 3 ) = { T = [ R t 0 T 1 ] ∈ R 4 × 4 ∣ R ∈ S O ( 3 ) , t ∈ R 3 } SE(3)=\lbrace{T=\begin{bmatrix} R & t \\ 0^T & 1 \end{bmatrix}\isin\R^{4\times4}| R\isin SO(3),t\isin \R^3}\rbrace SE(3)={T=[R0Tt1]R4×4RSO(3),tR3};
    验证封闭性:设 T = T 1 T 2 T=T_1T_2 T=T1T2,故 T = [ R 1 R 2 R 1 t 2 + t 1 0 T 1 ] T=\begin{bmatrix} R_1R_2 & R_1t_2+t_1 \\ 0^T & 1 \end{bmatrix} T=[R1R20TR1t2+t11],由于 R 1 R 2 ∈ S O ( 3 ) R_1R_2\isin SO(3) R1R2SO(3),且 R 1 t 2 + t 1 ∈ R 3 R_1t_2+t_1\isin R^3 R1t2+t1R3,因此 T ∈ S E ( 3 ) T\isin SE(3) TSE(3)
    验证结合律:根据矩阵乘法结合律,不难得到 ( T 1 T 2 ) T 3 = T 1 ( T 2 T 3 ) (T_1T_2)T_3=T_1(T_2T_3) (T1T2)T3=T1(T2T3)
    幺元:根据单位矩阵的特性,可以发现 I T = T I = T IT=TI=T IT=TI=T
    逆:对于 T T T,必然存在 T i n v = [ R T − R T t 0 T 1 ] T_inv=\begin{bmatrix} R^T & -R^Tt \\ 0^T & 1 \end{bmatrix} Tinv=[RT0TRTt1]使得 T T i n v = I TT_inv=I TTinv=I
    由于 S E ( 3 ) SE(3) SE(3) S O ( 3 ) SO(3) SO(3)并无明显区别,此处不再验证。

  2. 验证 ( R 3 , R , × ) (\R^3,\R,\times) (R3,R,×)构成李代数。
    证:
    假设 X , Y , Z ∈ R 3 X,Y,Z\isin \R^3 X,Y,ZR3
    封闭性:不难发现 [ X , Y ] = X ∧ Y ∈ R 3 [X,Y]=X^{\land}Y\isin \R^3 [X,Y]=XYR3
    双线性: [ a X + b Y , Z ] = ( a X + b Y ) Z = a X ∧ Z + b Y ∧ Z = a [ X , Z ] + b [ Y , Z ] [aX+bY,Z]=(aX+bY)Z=aX^{\land}Z+bY^{\land}Z=a[X,Z]+b[Y,Z] [aX+bY,Z]=(aX+bY)Z=aXZ+bYZ=a[X,Z]+b[Y,Z];同理 [ Z , a X + b Y ] = Z ( a X + b Z ) = a Z ∧ X + b Z ∧ Y = a [ Z , X ] + b [ Z , Y ] [Z,aX+bY]=Z(aX+bZ)=aZ^{\land}X+bZ^{\land}Y=a[Z,X]+b[Z,Y] [ZaX+bY]=Z(aX+bZ)=aZX+bZY=a[Z,X]+b[Z,Y]
    自反性:根据向量运行,明显 [ X , X ] = X ∧ X = 0 [X,X]=X^{\land}X=0 [X,X]=XX=0;
    雅克比等价:由于 [ X , [ Y , Z ] ] + [ Z , [ X , Y ] ] + [ Y , [ Z , X ] ] = [ X , Y ∧ Z ] + [ Z , X ∧ Y ] + [ Y , Z ∧ X ] = X ∧ ( Y ∧ Z ) + Z ∧ ( X ∧ Y ) + Y ∧ ( Z ∧ X ) [X,[Y,Z]]+[Z,[X,Y]]+[Y,[Z,X]]=[X,Y^{\land}Z]+[Z,X^{\land}Y]+[Y,Z^{\land}X]=X^{\land}(Y^{\land}Z)+Z^{\land}(X^{\land}Y)+Y^{\land}(Z^{\land}X) [X,[Y,Z]]+[Z,[X,Y]]+[Y,[Z,X]]=[X,YZ]+[Z,XY]+[Y,ZX]=X(YZ)+Z(XY)+Y(ZX)
    根据三重积展开公式可以得到
    X ∧ ( Y ∧ Z ) = Y ( X ∙ Z ) − Z ( X ∙ Y ) X^{\land}(Y^{\land}Z)=Y(X\bullet Z)-Z(X\bullet Y) X(YZ)=Y(XZ)Z(XY)
    Z ∧ ( X ∧ Y ) = X ( Z ∙ Y ) − Y ( Z ∙ X ) Z^{\land}(X^{\land}Y)=X(Z\bullet Y)-Y(Z\bullet X) Z(XY)=X(ZY)Y(ZX)
    Y ∧ ( Z ∧ X ) = Z ( Y ∙ Z ) − X ( Y ∙ Z ) Y^{\land}(Z^{\land}X)=Z(Y\bullet Z)-X(Y\bullet Z) Y(ZX)=Z(YZ)X(YZ)
    上式相加为 0 0 0,得证。

  3. s o ( 3 ) so(3) so(3) s e ( 3 ) se(3) se(3)满足李代数要求的性质。 验证 s o ( 3 ) so(3) so(3) s e ( 3 ) se(3) se(3)满足李代数要求的性质。
    证:
    s o ( 3 ) so(3) so(3)的李括号计算为 [ ϕ 1 , ϕ 2 ] = ( Φ 1 Φ 2 − Φ 2 Φ 1 ) ∨ [\phi_1,\phi_2]=(\Phi_1\Phi_2-\Phi2\Phi1)^{\lor} [ϕ1,ϕ2]=(Φ1Φ2Φ2Φ1),其中 Φ \Phi Φ对应的是 ϕ \phi ϕ的反对称矩阵, ϕ 1 = ( a , b , c ) T \phi_1=(a,b,c)^T ϕ1=(a,b,c)T ϕ 2 = ( d , e , f ) T \phi_2=(d,e,f)^T ϕ2=(d,e,f)T
    封闭性: ( Φ 1 Φ 2 − Φ 2 Φ 1 ) = [ 0 − c b c 0 − a − b a 0 ] [ 0 − f e f 0 − d − e d 0 ] − [ 0 − f e f 0 − d − e d 0 ] [ 0 − c b c 0 − a − b a 0 ] = [ − c f − b e b d d c a e − c f − a d c e a f b f − b e − a d ] − [ − c f − b e a e a f b d − c f − a d b f d c c e − b e − a d ] = [ 0 b d − a e d c − a f a e − b d − 0 c e − b f a f − d c b f − c e 0 ] (\Phi_1\Phi_2-\Phi2\Phi1)= \begin{bmatrix} 0 & -c & b\\ c & 0 & -a \\ -b & a & 0 \end{bmatrix}\begin{bmatrix} 0 & -f & e\\ f & 0 & -d \\ -e & d & 0 \end{bmatrix}-\begin{bmatrix} 0 & -f & e\\ f & 0 & -d \\ -e & d & 0 \end{bmatrix}\begin{bmatrix} 0 & -c & b\\ c & 0 & -a \\ -b & a & 0 \end{bmatrix}=\begin{bmatrix} -cf-be & bd & dc\\ ae & -cf-ad & ce \\ af & bf & -be-ad \end{bmatrix}-\begin{bmatrix} -cf-be & ae & af\\ bd & -cf-ad & bf\\ dc & ce& -be-ad \end{bmatrix}=\begin{bmatrix} 0 & bd-ae & dc-af\\ ae-bd & -0 & ce-bf \\ af-dc & bf-ce & 0 \end{bmatrix} (Φ1Φ2Φ2Φ1)=0cbc0aba00fef0ded00fef0ded00cbc0aba0=cfbeaeafbdcfadbfdccebeadcfbebddcaecfadceafbfbead=0aebdafdcbdae0bfcedcafcebf0,可以发现李括号运算后仍旧是对称矩形,即可以通过 ∨ \lor 将其转换为三维向量;
    双线性: [ a ϕ 1 + b ϕ 2 , ϕ 3 ] = ( ( a ϕ 1 + b ϕ 2 ) ∧ Φ 3 − Φ 3 ( a ϕ 1 + b ϕ 2 ) ∧ ) ∨ [a\phi_1+b\phi_2,\phi_3]=((a\phi_1+b\phi_2)^{ \land}\Phi_3-\Phi_3(a\phi_1+b\phi_2)^{\land})^{\lor} [aϕ1+bϕ2,ϕ3]=((aϕ1+bϕ2)Φ3Φ3(aϕ1+bϕ2)) ∧ \land 运算明显满足 ( ϕ 1 + ϕ 2 ) ∧ = ϕ 1 ∧ + ϕ 2 ∧ (\phi_1+\phi_2)^{\land}=\phi_1^{\land}+\phi_2^{\land} (ϕ1+ϕ2)=ϕ1+ϕ2;因此上式可以整理得到 ( ( a ϕ 1 + b ϕ 2 ) ∧ Φ 3 − Φ 3 ( a ϕ 1 + b ϕ 2 ) ∧ ) ∨ = ( a Φ 1 Φ 3 − a Φ 3 Φ 1 + b Φ 2 Φ 3 − b Φ 3 Φ 2 ) ∨ = a [ ϕ 1 , ϕ 3 ] + b [ ϕ 2 , ϕ 3 ] ((a\phi_1+b\phi_2)^{ \land}\Phi_3-\Phi_3(a\phi_1+b\phi_2)^{\land})^{\lor}=(a\Phi_1\Phi_3-a\Phi_3\Phi_1+b\Phi_2\Phi_3-b\Phi_3\Phi2)^{\lor}=a[\phi_1,\phi_3]+b[\phi_2,\phi_3] ((aϕ1+bϕ2)Φ3Φ3(aϕ1+bϕ2))=(aΦ1Φ3aΦ3Φ1+bΦ2Φ3bΦ3Φ2)=a[ϕ1,ϕ3]+b[ϕ2,ϕ3],得证;
    自反性:明显 [ ϕ , ϕ ] = ( Φ Φ − Φ Φ ) ∨ = 0 [\phi,\phi]=(\Phi\Phi-\Phi\Phi)^{\lor}=0 [ϕ,ϕ]=(ΦΦΦΦ)=0
    雅克比等价: [ ϕ 1 , [ ϕ 2 , ϕ 3 ] ] = [ ϕ 1 , ( Φ 2 Φ 3 − Φ 3 Φ 2 ) ∨ ] = ( Φ 1 ( Φ 2 Φ 3 − Φ 3 Φ 2 ) − ( Φ 2 Φ 3 − Φ 3 Φ 2 ) Φ 1 ) ∨ = ( Φ 1 Φ 2 Φ 3 − Φ 1 Φ 3 Φ 2 − Φ 2 Φ 3 Φ 1 + Φ 3 Φ 2 Φ 1 ) ∨ [\phi_1,[\phi_2,\phi_3]]=[\phi_1,(\Phi_2\Phi_3-\Phi_3\Phi_2)^{\lor}]=(\Phi_1(\Phi_2\Phi_3-\Phi_3\Phi_2)-(\Phi_2\Phi_3-\Phi_3\Phi_2)\Phi_1)^{\lor}=(\Phi_1\Phi_2\Phi_3-\Phi_1\Phi_3\Phi_2-\Phi_2\Phi_3\Phi_1+\Phi_3\Phi_2\Phi_1)^{\lor} [ϕ1,[ϕ2,ϕ3]]=[ϕ1,(Φ2Φ3Φ3Φ2)]=(Φ1(Φ2Φ3Φ3Φ2)(Φ2Φ3Φ3Φ2)Φ1)=(Φ1Φ2Φ3Φ1Φ3Φ2Φ2Φ3Φ1+Φ3Φ2Φ1)
    [ ϕ 3 , [ ϕ 1 , ϕ 2 ] ] = ( Φ 3 Φ 1 Φ 2 − Φ 3 Φ 2 Φ 1 − Φ 1 Φ 2 Φ 3 + Φ 2 Φ 1 Φ 3 ) ∨ [\phi_3,[\phi_1,\phi_2]]=(\Phi_3\Phi_1\Phi_2-\Phi_3\Phi_2\Phi_1-\Phi_1\Phi_2\Phi_3+\Phi_2\Phi_1\Phi_3)^{\lor} [ϕ3,[ϕ1,ϕ2]]=(Φ3Φ1Φ2Φ3Φ2Φ1Φ1Φ2Φ3+Φ2Φ1Φ3)
    [ ϕ 2 , [ ϕ 3 , ϕ 1 ] ] = ( Φ 2 Φ 3 Φ 1 − Φ 2 Φ 1 Φ 3 − Φ 3 Φ 1 Φ 2 + Φ 1 Φ 3 Φ 2 ) ∨ [\phi_2,[\phi_3,\phi_1]]=(\Phi_2\Phi_3\Phi_1-\Phi_2\Phi_1\Phi_3-\Phi_3\Phi_1\Phi_2+\Phi_1\Phi_3\Phi_2)^{\lor} [ϕ2,[ϕ3,ϕ1]]=(Φ2Φ3Φ1Φ2Φ1Φ3Φ3Φ1Φ2+Φ1Φ3Φ2)
    [ ϕ 1 , [ ϕ 2 , ϕ 3 ] ] + [ ϕ 3 , [ ϕ 1 , ϕ 2 ] ] + [ ϕ 2 , [ ϕ 3 , ϕ 1 ] ] = ( Φ 1 Φ 2 Φ 3 − Φ 1 Φ 3 Φ 2 − Φ 2 Φ 3 Φ 1 + Φ 3 Φ 2 Φ 1 + Φ 3 Φ 1 Φ 2 − Φ 3 Φ 2 Φ 1 − Φ 1 Φ 2 Φ 3 + Φ 2 Φ 1 Φ 3 + Φ 2 Φ 3 Φ 1 − Φ 2 Φ 1 Φ 3 − Φ 3 Φ 1 Φ 2 + Φ 1 Φ 3 Φ 2 ) ∨ = 0 [\phi_1,[\phi_2,\phi_3]]+[\phi_3,[\phi_1,\phi_2]]+[\phi_2,[\phi_3,\phi_1]]=(\Phi_1\Phi_2\Phi_3-\Phi_1\Phi_3\Phi_2-\Phi_2\Phi_3\Phi_1+\Phi_3\Phi_2\Phi_1+\Phi_3\Phi_1\Phi_2-\Phi_3\Phi_2\Phi_1-\Phi_1\Phi_2\Phi_3+\Phi_2\Phi_1\Phi_3+\Phi_2\Phi_3\Phi_1-\Phi_2\Phi_1\Phi_3-\Phi_3\Phi_1\Phi_2+\Phi_1\Phi_3\Phi_2)^{\lor}=0 [ϕ1,[ϕ2,ϕ3]]+[ϕ3,[ϕ1,ϕ2]]+[ϕ2,[ϕ3,ϕ1]]=(Φ1Φ2Φ3Φ1Φ3Φ2Φ2Φ3Φ1+Φ3Φ2Φ1+Φ3Φ1Φ2Φ3Φ2Φ1Φ1Φ2Φ3+Φ2Φ1Φ3+Φ2Φ3Φ1Φ2Φ1Φ3Φ3Φ1Φ2+Φ1Φ3Φ2)=0

    s e ( 3 ) se(3) se(3)的李括号计算为 [ ε 1 , ε 2 ] = ( ε 1 ∧ ε 2 ∧ − ε 2 ∧ ε 1 ∧ ) ∨ [\varepsilon_1,\varepsilon_2]=(\varepsilon_1^{\land}\varepsilon_2^{\land}-\varepsilon_2^{\land}\varepsilon_1^{\land})^{\lor} [ε1,ε2]=(ε1ε2ε2ε1),其中 ε ∧ = [ ϕ ∧ ρ 0 T 0 ] \varepsilon^{\land}=\begin{bmatrix} \phi^{\land} & \rho \\ 0^T & 0 \end{bmatrix} ε=[ϕ0Tρ0];
    封闭性: [ ε 1 , ε 2 ] = ( [ Φ 1 ρ 1 0 T 0 ] [ Φ 2 ρ 2 0 T 0 ] − [ Φ 2 ρ 2 0 T 0 ] [ Φ 1 ρ 1 0 T 0 ] ) ∨ = ( [ Φ 1 Φ 2 Φ 1 ρ 2 0 T 0 ] − [ Φ 2 Φ 1 Φ 2 ρ 1 0 T 1 ] ) ∨ = ( [ Φ 1 Φ 2 − Φ 2 Φ 1 Φ 1 ρ 2 − Φ 2 ρ 1 0 T 0 ] ) ∨ [\varepsilon_1,\varepsilon_2]=(\begin{bmatrix} \Phi_1 & \rho_1 \\ 0^T & 0 \end{bmatrix}\begin{bmatrix} \Phi_2 & \rho_2 \\ 0^T & 0 \end{bmatrix}-\begin{bmatrix} \Phi_2 & \rho_2 \\ 0^T & 0 \end{bmatrix}\begin{bmatrix} \Phi_1 & \rho_1 \\ 0^T & 0 \end{bmatrix})^{\lor}=(\begin{bmatrix} \Phi_1\Phi_2 & \Phi_1\rho_2 \\ 0^T & 0 \end{bmatrix}-\begin{bmatrix} \Phi_2\Phi_1 & \Phi_2\rho_1 \\ 0^T & 1 \end{bmatrix})^{\lor}=(\begin{bmatrix} \Phi_1\Phi_2-\Phi_2\Phi_1 & \Phi_1\rho_2-\Phi_2\rho_1 \\ 0^T & 0 \end{bmatrix})^{\lor} [ε1,ε2]=([Φ10Tρ10][Φ20Tρ20][Φ20Tρ20][Φ10Tρ10])=([Φ1Φ20TΦ1ρ20][Φ2Φ10TΦ2ρ11])=([Φ1Φ2Φ2Φ10TΦ1ρ2Φ2ρ10]),可以证明计算后得到的值属于 s e ( 3 ) se(3) se(3)
    双线性: [ a ε 1 + b ε 2 , ε 3 ] = ( ( a ε 1 + b ε 2 ) ∧ ε 3 ∧ − ε 3 ∧ ( a ε 1 + b ε 2 ) ∧ ) ∨ [a\varepsilon_1+b\varepsilon_2,\varepsilon_3]=((a\varepsilon_1+b\varepsilon_2)^{\land}\varepsilon_3^{\land}-\varepsilon_3^{\land}(a\varepsilon_1+b\varepsilon_2)^{\land})^{\lor} [aε1+bε2,ε3]=((aε1+bε2)ε3ε3(aε1+bε2)),虽然此处的 ∧ \land 非常复杂,但是明显也存在 ( ε 1 + ε 2 ) ∧ = ε 1 ∧ + ε 2 ∧ (\varepsilon_1+\varepsilon_2)^{\land}=\varepsilon_1{\land}+\varepsilon_2{\land} (ε1+ε2)=ε1+ε2(我也是猜想~),因此上式就可以类似 s o ( 3 ) so(3) so(3)展开证明;
    自反性: [ ε , ε ] = ( ε ∧ ε ∧ − ε ∧ ε ∧ ) ∨ = 0 [\varepsilon,\varepsilon]=(\varepsilon^{\land}\varepsilon^{\land}-\varepsilon^{\land}\varepsilon^{\land})^{\lor}=0 [ε,ε]=(εεεε)=0
    雅克比等价: [ ε 1 , [ ε 2 , ε 3 ] ] = [ ε 1 , ( ε 2 ∧ ε 3 ∧ − ε 3 ∧ ε 2 ∧ ) ∨ ] = ( ε 1 ∧ ( ε 2 ∧ ε 3 ∧ ) − ( ε 2 ∧ ε 3 ∧ ) ε 1 ∧ ) ∨ [\varepsilon_1,[\varepsilon_2,\varepsilon_3]]=[\varepsilon_1,(\varepsilon_2^{\land}\varepsilon_3^{\land}-\varepsilon_3^{\land}\varepsilon_2^{\land})^{\lor}]=(\varepsilon_1^{\land}(\varepsilon_2^{\land}\varepsilon_3^{\land})-(\varepsilon_2^{\land}\varepsilon_3^{\land})\varepsilon_1^{\land})^{\lor} [ε1,[ε2,ε3]]=[ε1,(ε2ε3ε3ε2)]=(ε1(ε2ε3)(ε2ε3)ε1)
    [ ε 3 , [ ε 1 , ε 2 ] ] = ( ε 3 ∧ ( ε 1 ∧ ε 2 ∧ ) − ( ε 1 ∧ ε 2 ∧ ) ε 3 ∧ ) ∨ [\varepsilon_3,[\varepsilon_1,\varepsilon_2]]=(\varepsilon_3^{\land}(\varepsilon_1^{\land}\varepsilon_2^{\land})-(\varepsilon_1^{\land}\varepsilon_2^{\land})\varepsilon_3^{\land})^{\lor} [ε3,[ε1,ε2]]=(ε3(ε1ε2)(ε1ε2)ε3)
    [ ε 2 , [ ε 3 , ε 1 ] ] = ( ε 2 ∧ ( ε 3 ∧ ε 1 ∧ ) − ( ε 3 ∧ ε 1 ∧ ) ε 2 ∧ ) ∨ [\varepsilon_2,[\varepsilon_3,\varepsilon_1]]=(\varepsilon_2^{\land}(\varepsilon_3^{\land}\varepsilon_1^{\land})-(\varepsilon_3^{\land}\varepsilon_1^{\land})\varepsilon_2^{\land})^{\lor} [ε2,[ε3,ε1]]=(ε2(ε3ε1)(ε3ε1)ε2)
    三者相加就可证明。

  4. 证明性质 ( 4.20 ) (4.20) (4.20) ( 4.21 ) (4.21) (4.21)
    证:
    首先记 a = ( x , y , z ) T a=(x,y,z)^T a=(x,y,z)T,且 x 2 + y 2 + z 2 = 1 x^2+y^2+z^2=1 x2+y2+z2=1
    ( 4.20 ) (4.20) (4.20) a ∧ a ∧ = a a T − I a^{\land}a^{\land}=aa^T-I aa=aaTI,,
    a ∧ a ∧ = [ 0 − z y z 0 − x − y x 0 ] [ 0 − z y z 0 − x − y x 0 ] = [ − y 2 − z 2 x y x z x y − z 2 − x 2 y z x z y z − x 2 − y 2 ] a^{\land}a^{\land}=\begin{bmatrix} 0 & -z & y \\ z & 0 & -x \\ -y & x & 0 \end{bmatrix}\begin{bmatrix} 0 & -z & y \\ z & 0 & -x \\ -y & x & 0 \end{bmatrix}=\begin{bmatrix} -y^2-z^2 & xy & xz \\ xy & -z^2-x^2 & yz \\ xz & yz & -x^2-y^2 \end{bmatrix} aa=0zyz0xyx00zyz0xyx0=y2z2xyxzxyz2x2yzxzyzx2y2
    a a T − I = [ x y z ] [ x y z ] − I = [ x 2 x y x z x y y 2 y z x z y z z 2 ] − I = [ x 2 − 1 x y x z x y y 2 − 1 y z x z y z z 2 − 1 ] = [ − y 2 − z 2 x y x z x y − z 2 − x 2 y z x z y z − x 2 − y 2 ] aa^T-I=\begin{bmatrix} x \\ y \\ z \end{bmatrix}\begin{bmatrix} x & y & z \end{bmatrix}-I=\begin{bmatrix} x^2 & xy & xz \\ xy & y^2 & yz \\ xz & yz & z^2 \end{bmatrix}-I=\begin{bmatrix} x^2-1 & xy & xz \\ xy & y^2-1 & yz \\ xz & yz & z^2-1 \end{bmatrix}=\begin{bmatrix} -y^2-z^2 & xy & xz \\ xy & -z^2-x^2 & yz \\ xz & yz & -x^2-y^2 \end{bmatrix} aaTI=xyz[xyz]I=x2xyxzxyy2yzxzyzz2I=x21xyxzxyy21yzxzyzz21=y2z2xyxzxyz2x2yzxzyzx2y2
    故得证。

    ( 4.21 ) (4.21) (4.21) a ∧ a ∧ a ∧ = − a ∧ a^{\land}a^{\land}a^{\land}=-a^{\land} aaa=a,由于 a ∧ a ∧ = [ − y 2 − z 2 x y x z x y − z 2 − x 2 y z x z y z − x 2 − y 2 ] a^{\land}a^{\land}=\begin{bmatrix} -y^2-z^2 & xy & xz \\ xy & -z^2-x^2 & yz \\ xz & yz & -x^2-y^2 \end{bmatrix} aa=y2z2xyxzxyz2x2yzxzyzx2y2,故 a ∧ a ∧ a ∧ = [ − y 2 − z 2 x y x z x y − z 2 − x 2 y z x z y z − x 2 − y 2 ] [ 0 − z y z 0 − x − y x 0 ] = [ 0 z ( y 2 + z 2 ) + z x 2 − y ( y 2 + z 2 ) − y x 2 z ( − z 2 − x 2 ) − z y 2 0 x y 2 + x ( z 2 + x 2 ) y z 2 + y ( x 2 + y 2 ) − x z 2 − x ( x 2 + y 2 ) 0 ] = [ 0 z − y − z 0 x y − x 0 ] = − a ∧ a^{\land}a^{\land}a^{\land}=\begin{bmatrix} -y^2-z^2 & xy & xz \\ xy & -z^2-x^2 & yz \\ xz & yz & -x^2-y^2 \end{bmatrix}\begin{bmatrix} 0 & -z & y \\ z & 0 & -x \\ -y & x & 0 \end{bmatrix}=\begin{bmatrix} 0 & z(y^2+z^2)+zx^2 & -y(y^2+z^2)-yx^2 \\ z(-z^2-x^2)-zy^2 & 0 & xy^2+x(z^2+x^2) \\ yz^2+y(x^2+y^2) & -xz^2-x(x^2+y^2) & 0 \end{bmatrix}=\begin{bmatrix} 0 & z & -y \\ -z & 0 & x \\ y & -x & 0 \end{bmatrix}=-a^{\land} aaa=y2z2xyxzxyz2x2yzxzyzx2y20zyz0xyx0=0z(z2x2)zy2yz2+y(x2+y2)z(y2+z2)+zx20xz2x(x2+y2)y(y2+z2)yx2xy2+x(z2+x2)0=0zyz0xyx0=a

你可能感兴趣的:(SLAM)