【控制】多智能体系统总结。1. 系统模型。2.控制目标。3.模型转换。
【控制】多智能体系统总结。4.控制协议。
【控制】多智能体系统总结。5.系统合并。

4. 控制协议

4.1 一阶一维系统

u i = α ∑ j ∈ N i a i j ( p j − p i ) ( ) u_i = \alpha \sum_{j \in N_i} a_{ij} (p_j - p_i) \tag{} ui=αjNiaij(pjpi)()

u 1 = α a 11 ( p 1 − p 1 ) + α a 12 ( p 2 − p 1 ) + α a 13 ( p 3 − p 1 ) + ⋯ u 2 = α a 21 ( p 1 − p 2 ) + α a 22 ( p 2 − p 2 ) + α a 23 ( p 3 − p 2 ) + ⋯ u 3 = α a 31 ( p 1 − p 3 ) + α a 32 ( p 2 − p 3 ) + α a 33 ( p 3 − p 3 ) + ⋯ ( ) \begin{aligned} u_1 &= {\alpha} a_{11} (p_1 - p_1) &+ {\alpha} a_{12} (p_2 - p_1) &+ {\alpha} a_{13} (p_3 - p_1) &+ \cdots \\ u_2 &= {\alpha} a_{21} (p_1 - p_2) &+ {\alpha} a_{22} (p_2 - p_2) &+ {\alpha} a_{23} (p_3 - p_2) &+ \cdots \\ u_3 &= {\alpha} a_{31} (p_1 - p_3) &+ {\alpha} a_{32} (p_2 - p_3) &+ {\alpha} a_{33} (p_3 - p_3) &+ \cdots \\ \end{aligned} \tag{} u1u2u3=αa11(p1p1)=αa21(p1p2)=αa31(p1p3)+αa12(p2p1)+αa22(p2p2)+αa32(p2p3)+αa13(p3p1)+αa23(p3p2)+αa33(p3p3)+++()

[ u 1 u 2 u 3 ] = [ − α d 1 α a 12 α a 13 α a 21 − α d 2 α a 23 α a 31 α a 32 − α d 3 ] [ p 1 p 2 p 3 ] = − α L ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1 \\ u_2 \\ u_3 \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & \alpha a_{12} & \alpha a_{13} \\ \alpha a_{21} & -\alpha d_{ 2} & \alpha a_{23} \\ \alpha a_{31} & \alpha a_{32} & -\alpha d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1 \\ p_2 \\ p_3 \\ \end{matrix}\right] \\ &= \red{-\alpha L \cdot X} \end{aligned} \tag{} u1u2u3=αd1αa21αa31αa12αd2αa32αa13αa23αd3p1p2p3=αLX()

4.2 一阶二维系统

u i x = α ∑ j ∈ N i a i j ( p j x − p i x ) u i y = α ∑ j ∈ N i a i j ( p j y − p i y ) ( ) \begin{aligned} u_i^x &= \alpha \sum_{j \in N_i} a_{ij} (p_j^x - p_i^x) \\ u_i^y &= \alpha \sum_{j \in N_i} a_{ij} (p_j^y - p_i^y) \\ \end{aligned}\tag{} uixuiy=αjNiaij(pjxpix)=αjNiaij(pjypiy)()

u 1 x = α a 11 ( p 1 x − p 1 x ) + α a 12 ( p 2 x − p 1 x ) + α a 13 ( p 3 x − p 1 x ) + ⋯ u 1 y = α a 11 ( p 1 y − p 1 y ) + α a 12 ( p 2 y − p 1 y ) + α a 13 ( p 3 y − p 1 y ) + ⋯ u 2 x = α a 21 ( p 1 x − p 2 x ) + α a 22 ( p 2 x − p 2 x ) + α a 23 ( p 3 x − p 2 x ) + ⋯ u 2 y = α a 21 ( p 1 y − p 2 y ) + α a 22 ( p 2 y − p 2 y ) + α a 23 ( p 3 y − p 2 y ) + ⋯ u 3 x = α a 31 ( p 1 x − p 3 x ) + α a 32 ( p 2 x − p 3 x ) + α a 33 ( p 3 x − p 3 x ) + ⋯ u 3 y = α a 31 ( p 1 y − p 3 y ) + α a 32 ( p 2 y − p 3 y ) + α a 33 ( p 3 y − p 3 y ) + ⋯ ( ) \begin{aligned} u_1^x &= {\alpha} a_{11} (p_1^x - p_1^x) &+ {\alpha} a_{12} (p_2^x - p_1^x) &+ {\alpha} a_{13} (p_3^x - p_1^x) &+ \cdots \\ u_1^y &= {\alpha} a_{11} (p_1^y - p_1^y) &+ {\alpha} a_{12} (p_2^y - p_1^y) &+ {\alpha} a_{13} (p_3^y - p_1^y) &+ \cdots \\ u_2^x &= {\alpha} a_{21} (p_1^x - p_2^x) &+ {\alpha} a_{22} (p_2^x - p_2^x) &+ {\alpha} a_{23} (p_3^x - p_2^x) &+ \cdots \\ u_2^y &= {\alpha} a_{21} (p_1^y - p_2^y) &+ {\alpha} a_{22} (p_2^y - p_2^y) &+ {\alpha} a_{23} (p_3^y - p_2^y) &+ \cdots \\ u_3^x &= {\alpha} a_{31} (p_1^x - p_3^x) &+ {\alpha} a_{32} (p_2^x - p_3^x) &+ {\alpha} a_{33} (p_3^x - p_3^x) &+ \cdots \\ u_3^y &= {\alpha} a_{31} (p_1^y - p_3^y) &+ {\alpha} a_{32} (p_2^y - p_3^y) &+ {\alpha} a_{33} (p_3^y - p_3^y) &+ \cdots \\ \end{aligned} \tag{} u1xu1yu2xu2yu3xu3y=αa11(p1xp1x)=αa11(p1yp1y)=αa21(p1xp2x)=αa21(p1yp2y)=αa31(p1xp3x)=αa31(p1yp3y)+αa12(p2xp1x)+αa12(p2yp1y)+αa22(p2xp2x)+αa22(p2yp2y)+αa32(p2xp3x)+αa32(p2yp3y)+αa13(p3xp1x)+αa13(p3yp1y)+αa23(p3xp2x)+αa23(p3yp2y)+αa33(p3xp3x)+αa33(p3yp3y)++++++()

4.2.1 方式一

[ u 1 x u 1 y u 2 x u 2 y u 3 x u 3 y ] = [ − α d 1 0 α a 12 0 α a 13 0 0 − α d 1 0 α a 12 0 α a 13 α a 21 0 − α d 2 0 α a 23 0 0 α a 21 0 − α d 2 0 α a 23 α a 31 0 α a 32 0 − α d 3 0 0 α a 31 0 α a 32 0 − α d 3 ] [ p 1 x p 1 y p 2 x p 2 y p 3 x p 3 y ] = − α L ⊗ I 2 ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1^x \\ u_1^y \\ u_2^x \\ u_2^y \\ u_3^x \\ u_3^y \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & 0 & \alpha a_{12} & 0 & \alpha a_{13} & 0 \\ 0 & -\alpha d_{ 1} & 0 & \alpha a_{12} & 0 & \alpha a_{13} \\ \alpha a_{21} & 0 & -\alpha d_{ 2} & 0 & \alpha a_{23} & 0 \\ 0 & \alpha a_{21} & 0 & -\alpha d_{ 2} & 0 & \alpha a_{23} \\ \alpha a_{31} & 0 & \alpha a_{32} & 0 & -\alpha d_{ 3} & 0 \\ 0 & \alpha a_{31} & 0 & \alpha a_{32} & 0 & -\alpha d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1^x \\ p_1^y \\ p_2^x \\ p_2^y \\ p_3^x \\ p_3^y \\ \end{matrix}\right] \\ &= \red{-\alpha L \otimes I_2 \cdot X} \end{aligned} \tag{} u1xu1yu2xu2yu3xu3y=αd10αa210αa3100αd10αa210αa31αa120αd20αa3200αa120αd20αa32αa130αa230αd300αa130αa230αd3p1xp1yp2xp2yp3xp3y=αLI2X()

4.2.2 方式二

[ u 1 x u 2 x u 3 x u 1 y u 2 y u 3 y ] = [ − α d 1 α a 12 α a 13 0 0 0 α a 21 − α d 2 α a 23 0 0 0 α a 31 α a 32 − α d 3 0 0 0 0 0 0 − α d 1 α a 12 α a 13 0 0 0 α a 21 − α d 2 α a 23 0 0 0 α a 31 α a 32 − α d 3 ] [ p 1 x p 2 x p 3 x p 1 y p 2 y p 3 y ] = I 2 ⊗ − α L ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1^x \\ u_2^x \\ u_3^x \\ u_1^y \\ u_2^y \\ u_3^y \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & \alpha a_{12} & \alpha a_{13} & 0 & 0 & 0 \\ \alpha a_{21} & -\alpha d_{ 2} & \alpha a_{23} & 0 & 0 & 0 \\ \alpha a_{31} & \alpha a_{32} & -\alpha d_{ 3} & 0 & 0 & 0 \\ 0 & 0 & 0 & -\alpha d_{ 1} & \alpha a_{12} & \alpha a_{13} \\ 0 & 0 & 0 & \alpha a_{21} & -\alpha d_{ 2} & \alpha a_{23} \\ 0 & 0 & 0 & \alpha a_{31} & \alpha a_{32} & -\alpha d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1^x \\ p_2^x \\ p_3^x \\ p_1^y \\ p_2^y \\ p_3^y \\ \end{matrix}\right] \\ &= \red{I_2 \otimes -\alpha L \cdot X} \end{aligned} \tag{} u1xu2xu3xu1yu2yu3y=αd1αa21αa31000αa12αd2αa32000αa13αa23αd3000000αd1αa21αa31000αa12αd2αa32000αa13αa23αd3p1xp2xp3xp1yp2yp3y=I2αLX()

4.3 二阶一维系统

u i = α ∑ j ∈ N i a i j ( p j − p i ) + β ∑ j ∈ N i a i j ( v j − v i ) ( ) u_i = \alpha \sum_{j \in N_i} a_{ij} (p_j - p_i) + \beta \sum_{j \in N_i} a_{ij} (v_j - v_i) \tag{} ui=αjNiaij(pjpi)+βjNiaij(vjvi)()

u 1 = α a 11 ( p 1 − p 1 ) + β a 11 ( v 1 − v 1 ) + α a 12 ( p 2 − p 1 ) + β a 12 ( v 2 − v 1 ) + α a 13 ( p 3 − p 1 ) + β a 13 ( v 3 − v 1 ) + ⋯ u 2 = α a 21 ( p 1 − p 2 ) + β a 21 ( v 1 − v 2 ) + α a 22 ( p 2 − p 2 ) + β a 22 ( v 2 − v 2 ) + α a 23 ( p 3 − p 2 ) + β a 23 ( v 3 − v 2 ) + ⋯ u 3 = α a 31 ( p 1 − p 3 ) + β a 31 ( v 1 − v 3 ) + α a 32 ( p 2 − p 3 ) + β a 32 ( v 2 − v 3 ) + α a 33 ( p 3 − p 3 ) + β a 33 ( v 3 − v 3 ) + ⋯ ( ) \begin{aligned} u_1 &= {\alpha} a_{11} (p_1 - p_1) + {\beta} a_{11} (v_1 - v_1) &+ {\alpha} a_{12} (p_2 - p_1) + {\beta} a_{12} (v_2 - v_1) &+ {\alpha} a_{13} (p_3 - p_1) + {\beta} a_{13} (v_3 - v_1) &+ \cdots \\ u_2 &= {\alpha} a_{21} (p_1 - p_2) + {\beta} a_{21} (v_1 - v_2) &+ {\alpha} a_{22} (p_2 - p_2) + {\beta} a_{22} (v_2 - v_2) &+ {\alpha} a_{23} (p_3 - p_2) + {\beta} a_{23} (v_3 - v_2) &+ \cdots \\ u_3 &= {\alpha} a_{31} (p_1 - p_3) + {\beta} a_{31} (v_1 - v_3) &+ {\alpha} a_{32} (p_2 - p_3) + {\beta} a_{32} (v_2 - v_3) &+ {\alpha} a_{33} (p_3 - p_3) + {\beta} a_{33} (v_3 - v_3) &+ \cdots \\ \end{aligned} \tag{} u1u2u3=αa11(p1p1)+βa11(v1v1)=αa21(p1p2)+βa21(v1v2)=αa31(p1p3)+βa31(v1v3)+αa12(p2p1)+βa12(v2v1)+αa22(p2p2)+βa22(v2v2)+αa32(p2p3)+βa32(v2v3)+αa13(p3p1)+βa13(v3v1)+αa23(p3p2)+βa23(v3v2)+αa33(p3p3)+βa33(v3v3)+++()

4.3.1 方式一

[ u 1 u 2 u 3 ] = [ − α d 1 − β d 1 α a 12 β a 12 α a 13 β a 13 α a 21 β a 21 − α d 2 − β d 2 α a 23 β a 23 α a 31 β a 31 α a 32 β a 32 − α d 3 − β d 3 ] [ p 1 v 1 p 2 v 2 p 3 v 3 ] = ( − L ) ⊗ [ α β ] ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1 \\ u_2 \\ u_3 \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & -\beta d_{ 1} & \alpha a_{12} & \beta a_{12} & \alpha a_{13} & \beta a_{13} \\ \alpha a_{21} & \beta a_{21} & -\alpha d_{ 2} & -\beta d_{ 2} & \alpha a_{23} & \beta a_{23} \\ \alpha a_{31} & \beta a_{31} & \alpha a_{32} & \beta a_{32} & -\alpha d_{ 3} & -\beta d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1 \\ v_1 \\ p_2 \\ v_2 \\ p_3 \\ v_3 \\ \end{matrix}\right] \\ &= \red{(-L) \otimes \left[\begin{matrix} \alpha & \beta \end{matrix}\right] \cdot X} \end{aligned} \tag{} u1u2u3=αd1αa21αa31βd1βa21βa31αa12αd2αa32βa12βd2βa32αa13αa23αd3βa13βa23βd3p1v1p2v2p3v3=(L)[αβ]X()

4.3.2 方式二

[ u 1 u 2 u 3 ] = [ − α d 1 α a 12 α a 13 − β d 1 β a 12 β a 13 α a 21 − α d 2 α a 23 β a 21 − β d 2 β a 23 α a 31 α a 32 − α d 3 β a 31 β a 32 − β d 3 ] [ p 1 p 2 p 3 v 1 v 2 v 3 ] = [ α β ] ⊗ ( − L ) ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1 \\ u_2 \\ u_3 \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & \alpha a_{12} & \alpha a_{13} & -\beta d_{ 1} & \beta a_{12} & \beta a_{13} \\ \alpha a_{21} & -\alpha d_{ 2} & \alpha a_{23} & \beta a_{21} & -\beta d_{ 2} & \beta a_{23} \\ \alpha a_{31} & \alpha a_{32} & -\alpha d_{ 3} & \beta a_{31} & \beta a_{32} & -\beta d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1 \\ p_2 \\ p_3 \\ v_1 \\ v_2 \\ v_3 \\ \end{matrix}\right] \\ &= \red{ \left[\begin{matrix} \alpha & \beta \end{matrix}\right] \otimes (-L) \cdot X} \end{aligned} \tag{} u1u2u3=αd1αa21αa31αa12αd2αa32αa13αa23αd3βd1βa21βa31βa12βd2βa32βa13βa23βd3p1p2p3v1v2v3=[αβ](L)X()

4.4 二阶二维系统

u i x = α ∑ j ∈ N i a i j ( p j x − p i x ) + β ∑ j ∈ N i a i j ( v j x − v i x ) u i y = α ∑ j ∈ N i a i j ( p j y − p i y ) + β ∑ j ∈ N i a i j ( v j y − v i y ) ( ) \begin{aligned} u_i^x &= \alpha \sum_{j \in N_i} a_{ij} (p_j^x - p_i^x) + \beta \sum_{j \in N_i} a_{ij} (v_j^x - v_i^x) \\ u_i^y &= \alpha \sum_{j \in N_i} a_{ij} (p_j^y - p_i^y) + \beta \sum_{j \in N_i} a_{ij} (v_j^y - v_i^y) \\ \end{aligned}\tag{} uixuiy=αjNiaij(pjxpix)+βjNiaij(vjxvix)=αjNiaij(pjypiy)+βjNiaij(vjyviy)()

u 1 = α a 11 ( p 1 x − p 1 x ) + β a 11 ( v 1 x − v 1 x ) + α a 12 ( p 2 x − p 1 x ) + β a 12 ( v 2 x − v 1 x ) + α a 13 ( p 3 x − p 1 x ) + β a 13 ( v 3 x − v 1 x ) + ⋯ u 2 = α a 21 ( p 1 − p 2 ) + β a 21 ( v 1 − v 2 ) + α a 22 ( p 2 − p 2 ) + β a 22 ( v 2 − v 2 ) + α a 23 ( p 3 − p 2 ) + β a 23 ( v 3 − v 2 ) + ⋯ u 3 = α a 31 ( p 1 − p 3 ) + β a 31 ( v 1 − v 3 ) + α a 32 ( p 2 − p 3 ) + β a 32 ( v 2 − v 3 ) + α a 33 ( p 3 − p 3 ) + β a 33 ( v 3 − v 3 ) + ⋯ ( ) \begin{aligned} u_1 &= {\alpha} a_{11} (p_1^x - p_1^x) + {\beta} a_{11} (v_1^x - v_1^x) &+ {\alpha} a_{12} (p_2^x - p_1^x) + {\beta} a_{12} (v_2^x - v_1^x) &+ {\alpha} a_{13} (p_3^x - p_1^x) + {\beta} a_{13} (v_3^x - v_1^x) &+ \cdots \\ u_2 &= {\alpha} a_{21} (p_1 - p_2) + {\beta} a_{21} (v_1 - v_2) &+ {\alpha} a_{22} (p_2 - p_2) + {\beta} a_{22} (v_2 - v_2) &+ {\alpha} a_{23} (p_3 - p_2) + {\beta} a_{23} (v_3 - v_2) &+ \cdots \\ u_3 &= {\alpha} a_{31} (p_1 - p_3) + {\beta} a_{31} (v_1 - v_3) &+ {\alpha} a_{32} (p_2 - p_3) + {\beta} a_{32} (v_2 - v_3) &+ {\alpha} a_{33} (p_3 - p_3) + {\beta} a_{33} (v_3 - v_3) &+ \cdots \\ \end{aligned} \tag{} u1u2u3=αa11(p1xp1x)+βa11(v1xv1x)=αa21(p1p2)+βa21(v1v2)=αa31(p1p3)+βa31(v1v3)+αa12(p2xp1x)+βa12(v2xv1x)+αa22(p2p2)+βa22(v2v2)+αa32(p2p3)+βa32(v2v3)+αa13(p3xp1x)+βa13(v3xv1x)+αa23(p3p2)+βa23(v3v2)+αa33(p3p3)+βa33(v3v3)+++()

4.4.1 方式一

[ u 1 x u 1 y u 2 x u 2 y u 3 x u 3 y ] = [ − α d 1 0 − β d 1 0 α a 12 0 β a 12 0 α a 13 0 β a 13 0 0 − α d 1 0 − β d 1 0 α a 12 0 β a 12 0 α a 13 0 β a 13 α a 21 0 β a 21 0 − α d 2 0 − β d 2 0 α a 23 0 β a 23 0 0 α a 21 0 β a 21 0 − α d 2 0 − β d 2 0 α a 23 0 β a 23 α a 31 0 β a 31 0 α a 32 0 β a 32 0 − α d 3 0 − β d 3 0 0 α a 31 0 β a 31 0 α a 32 0 β a 32 0 − α d 3 0 − β d 3 ] [ p 1 x p 1 y v 1 x v 1 y p 2 x p 2 y v 2 x v 2 y p 3 x p 3 y v 3 x v 3 y ] = − L ⊗ [ α β ] ⊗ I 2 ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1^x \\ u_1^y \\ u_2^x \\ u_2^y \\ u_3^x \\ u_3^y \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & 0 & -\beta d_{ 1} & 0 & \alpha a_{12} & 0 & \beta a_{12} & 0 & \alpha a_{13} & 0 & \beta a_{13} & 0 \\ 0 & -\alpha d_{ 1} & 0 & -\beta d_{ 1} & 0 & \alpha a_{12} & 0 & \beta a_{12} & 0 & \alpha a_{13} & 0 & \beta a_{13} \\ \alpha a_{21} & 0 & \beta a_{21} & 0 & -\alpha d_{ 2} & 0 & -\beta d_{ 2} & 0 & \alpha a_{23} & 0 & \beta a_{23} & 0 \\ 0 & \alpha a_{21} & 0 & \beta a_{21} & 0 & -\alpha d_{ 2} & 0 & -\beta d_{ 2} & 0 & \alpha a_{23} & 0 & \beta a_{23} \\ \alpha a_{31} & 0 & \beta a_{31} & 0 & \alpha a_{32} & 0 & \beta a_{32} & 0 & -\alpha d_{ 3} & 0 & -\beta d_{ 3} & 0 \\ 0 & \alpha a_{31} & 0 & \beta a_{31} & 0 & \alpha a_{32} & 0 & \beta a_{32} & 0 & -\alpha d_{ 3} & 0 & -\beta d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1^x \\ p_1^y \\ v_1^x \\ v_1^y \\ p_2^x \\ p_2^y \\ v_2^x \\ v_2^y \\ p_3^x \\ p_3^y \\ v_3^x \\ v_3^y \\ \end{matrix}\right] \\ &= \red{ -L \otimes \left[\begin{matrix} \alpha & \beta \end{matrix}\right] \otimes I_2 \cdot X} \end{aligned} \tag{} u1xu1yu2xu2yu3xu3y=αd10αa210αa3100αd10αa210αa31βd10βa210βa3100βd10βa210βa31αa120αd20αa3200αa120αd20αa32βa120βd20βa3200βa120βd20βa32αa130αa230αd300αa130αa230αd3βa130βa230βd300βa130βa230βd3p1xp1yv1xv1yp2xp2yv2xv2yp3xp3yv3xv3y=L[αβ]I2X()

4.4.2 方式二

[ u 1 x u 2 x u 3 x u 1 y u 2 y u 3 y ] = [ − α d 1 α a 12 α a 13 0 0 0 − β d 1 β a 12 β a 13 0 0 0 α a 21 − α d 2 α a 23 0 0 0 β a 21 − β d 2 β a 23 0 0 0 α a 31 α a 32 − α d 3 0 0 0 β a 31 β a 32 − β d 3 0 0 0 0 0 0 − α d 1 α a 12 α a 13 0 0 0 − β d 1 β a 12 β a 13 0 0 0 α a 21 − α d 2 α a 23 0 0 0 β a 21 − β d 2 β a 23 0 0 0 α a 31 α a 32 − α d 3 0 0 0 β a 31 β a 32 − β d 3 ] [ p 1 x p 2 x p 3 x p 1 y p 2 y p 3 y v 1 x v 2 x v 3 x v 1 y v 2 y v 3 y ] = [ α β ] ⊗ I 2 ⊗ ( − L ) ⋅ X ( ) \begin{aligned} \left[\begin{matrix} u_1^x \\ u_2^x \\ u_3^x \\ u_1^y \\ u_2^y \\ u_3^y \\ \end{matrix}\right] &= \left[\begin{matrix} -\alpha d_{ 1} & \alpha a_{12} & \alpha a_{13} & 0 & 0 & 0 &-\beta d_{ 1} & \beta a_{12} & \beta a_{13} & 0 & 0 & 0 \\ \alpha a_{21} & -\alpha d_{ 2} & \alpha a_{23} & 0 & 0 & 0 & \beta a_{21} & -\beta d_{ 2} & \beta a_{23} & 0 & 0 & 0 \\ \alpha a_{31} & \alpha a_{32} & -\alpha d_{ 3} & 0 & 0 & 0 & \beta a_{31} & \beta a_{32} & -\beta d_{ 3} & 0 & 0 & 0 \\ 0 & 0 & 0 & -\alpha d_{ 1} & \alpha a_{12} & \alpha a_{13} & 0 & 0 & 0 &-\beta d_{ 1} & \beta a_{12} & \beta a_{13} \\ 0 & 0 & 0 & \alpha a_{21} & -\alpha d_{ 2} & \alpha a_{23} & 0 & 0 & 0 & \beta a_{21} & -\beta d_{ 2} & \beta a_{23} \\ 0 & 0 & 0 & \alpha a_{31} & \alpha a_{32} & -\alpha d_{ 3} & 0 & 0 & 0 & \beta a_{31} & \beta a_{32} & -\beta d_{ 3} \\ \end{matrix}\right] \left[\begin{matrix} p_1^x \\ p_2^x \\ p_3^x \\ p_1^y \\ p_2^y \\ p_3^y \\ v_1^x \\ v_2^x \\ v_3^x \\ v_1^y \\ v_2^y \\ v_3^y \\ \end{matrix}\right] \\ &= \red{ \left[\begin{matrix} \alpha & \beta \end{matrix}\right] \otimes I_2 \otimes (-L) \cdot X} \end{aligned} \tag{} u1xu2xu3xu1yu2yu3y=αd1αa21αa31000αa12αd2αa32000αa13αa23αd3000000αd1αa21αa31000αa12αd2αa32000αa13αa23αd3βd1βa21βa31000βa12βd2βa32000βa13βa23βd3000000βd1βa21βa31000βa12βd2βa32000βa13βa23βd3p1xp2xp3xp1yp2yp3yv1xv2xv3xv1yv2yv3y=[αβ]I2(L)X()


3.1 动态一致性控制协议

控制协议为
u i = α ∑ j ∈ N i a i j ( p j − p i ) + β ∑ j ∈ N i a i j ( v j − v i ) ( ) u_i = \red{\alpha} \sum_{j \in N_i} a_{ij} (p_j - p_i) + \blue{\beta} \sum_{j \in N_i} a_{ij} (v_j - v_i) \tag{} ui=αjNiaij(pjpi)+βjNiaij(vjvi)()

其中
a i j a_{ij} aij:表示邻接矩阵
α \red{\alpha} α:表示控制参数
β \blue{\beta} β:表示控制参数


通过分析控制协议:
u 1 = α a 11 ( p 1 − p 1 ) + β a 11 ( v 1 − v 1 ) + α a 12 ( p 2 − p 1 ) + β a 12 ( v 2 − v 1 ) + α a 13 ( p 3 − p 1 ) + β a 13 ( v 3 − v 1 ) + ⋯ u 2 = α a 21 ( p 1 − p 2 ) + β a 21 ( v 1 − v 2 ) + α a 22 ( p 2 − p 2 ) + β a 22 ( v 2 − v 2 ) + α a 23 ( p 3 − p 2 ) + β a 23 ( v 3 − v 2 ) + ⋯ u 3 = α a 31 ( p 1 − p 3 ) + β a 31 ( v 1 − v 3 ) + α a 32 ( p 2 − p 3 ) + β a 32 ( v 2 − v 3 ) + α a 33 ( p 3 − p 3 ) + β a 33 ( v 3 − v 3 ) + ⋯ ( ) \begin{aligned} u_1 &= \red{\alpha} a_{11} (p_1 - p_1) + \blue{\beta} a_{11} (v_1 - v_1) &+ \red{\alpha} a_{12} (p_2 - p_1) + \blue{\beta} a_{12} (v_2 - v_1) &+ \red{\alpha} a_{13} (p_3 - p_1) + \blue{\beta} a_{13} (v_3 - v_1) &+ \cdots \\ u_2 &= \red{\alpha} a_{21} (p_1 - p_2) + \blue{\beta} a_{21} (v_1 - v_2) &+ \red{\alpha} a_{22} (p_2 - p_2) + \blue{\beta} a_{22} (v_2 - v_2) &+ \red{\alpha} a_{23} (p_3 - p_2) + \blue{\beta} a_{23} (v_3 - v_2) &+ \cdots \\ u_3 &= \red{\alpha} a_{31} (p_1 - p_3) + \blue{\beta} a_{31} (v_1 - v_3) &+ \red{\alpha} a_{32} (p_2 - p_3) + \blue{\beta} a_{32} (v_2 - v_3) &+ \red{\alpha} a_{33} (p_3 - p_3) + \blue{\beta} a_{33} (v_3 - v_3) &+ \cdots \\ \end{aligned} \tag{} u1u2u3=αa11(p1p1)+βa11(v1v1)=αa21(p1p2)+βa21(v1v2)=αa31(p1p3)+βa31(v1v3)+αa12(p2p1)+βa12(v2v1)+αa22(p2p2)+βa22(v2v2)+αa32(p2p3)+βa32(v2v3)+αa13(p3p1)+βa13(v3v1)+αa23(p3p2)+βa23(v3v2)+αa33(p3p3)+βa33(v3v3)+++()

可以得到以下控制协议的矩阵形式:

[ u 1 x u 1 y u 2 x u 2 y u 3 x u 3 y ] = [ − α d 1 0 − β d 1 0 α a 12 0 β a 12 0 α a 13 0 β a 13 0 0 − α d 1 0 − β d 1 0 α a 12 0 β a 12 0 α a 13 0 β a 13 α a 21 0 β a 21 0 − α d 2 0 − β d 2 0 α a 23 0 β a 23 0 0 α a 21 0 β a 21 0 − α d 2 0 − β d 2 0 α a 23 0 β a 23 α a 31 0 β a 31 0 α a 32 0 β a 32 0 − α d 3 0 − β d 3 0 0 α a 31 0 β a 31 0 α a 32 0 β a 32 0 − α d 3 0 − β d 3 ] [ p 1 x p 1 y v 1 x v 1 y p 2 x p 2 y v 2 x v 2 y p 3 x p 3 y v 3 x v 3 y ] = [ − d 1 a 12 a 13 a 21 − d 2 a 23 a 31 a 32 − d 3 ] ⊗ [ α β ] ⊗ I 2 ⋅ X i = − L ⊗ [ α β ] ⊗ I 2 ⋅ X i \begin{aligned} \left[\begin{matrix} u_1^x \\ u_1^y \\ u_2^x \\ u_2^y \\ u_3^x \\ u_3^y \\ \end{matrix}\right]&= \left[\begin{matrix} -\red{\alpha}d_1 & 0 & -\blue{\beta}d_1 & 0 & \red{\alpha}a_{12} & 0 & \blue{\beta}a_{12} & 0 & \red{\alpha}a_{13} & 0 & \blue{\beta}a_{13} & 0 \\ 0 & -\red{\alpha}d_1 & 0 & -\blue{\beta}d_1 & 0 & \red{\alpha}a_{12} & 0 & \blue{\beta}a_{12} & 0 & \red{\alpha}a_{13} & 0 & \blue{\beta}a_{13} \\ \red{\alpha}a_{21} & 0 & \blue{\beta}a_{21} & 0 & -\red{\alpha}d_{2} & 0 & -\blue{\beta}d_{2} & 0 & \red{\alpha}a_{23} & 0 & \blue{\beta}a_{23} & 0 \\ 0 & \red{\alpha}a_{21} & 0 & \blue{\beta}a_{21} & 0 & -\red{\alpha}d_{2} & 0 & -\blue{\beta}d_{2} & 0 & \red{\alpha}a_{23} & 0 & \blue{\beta}a_{23} \\ \red{\alpha}a_{31} & 0 & \blue{\beta}a_{31} & 0 & \red{\alpha}a_{32} & 0 & \blue{\beta}a_{32} & 0 & -\red{\alpha}d_{3} & 0 & -\blue{\beta}d_{3} & 0 \\ 0 & \red{\alpha}a_{31} & 0 & \blue{\beta}a_{31} & 0 & \red{\alpha}a_{32} & 0 & \blue{\beta}a_{32} & 0 & -\red{\alpha}d_{3} & 0 & -\blue{\beta}d_{3} \\ \end{matrix}\right] \left[\begin{matrix} p_1^x \\ p_1^y \\ v_1^x \\ v_1^y \\ p_2^x \\ p_2^y \\ v_2^x \\ v_2^y \\ p_3^x \\ p_3^y \\ v_3^x \\ v_3^y \\ \end{matrix}\right] \\ &= \left[\begin{matrix} -d_{1} & a_{12} & a_{13} \\ a_{21} & -d_{2} & a_{23} \\ a_{31} & a_{32} & -d_{3} \\ \end{matrix}\right] \otimes \left[\begin{matrix} \red{\alpha} & \blue{\beta} \end{matrix}\right] \otimes I_2 \cdot X_i \\ &= \red{-L \otimes \left[\begin{matrix} \red{\alpha} & \blue{\beta} \end{matrix}\right] \otimes I_2} \cdot X_i \end{aligned} u1xu1yu2xu2yu3xu3y=αd10αa210αa3100αd10αa210αa31βd10βa210βa3100βd10βa210βa31αa120αd20αa3200αa120αd20αa32βa120βd20βa3200βa120βd20βa32αa130αa230αd300αa130αa230αd3βa130βa230βd300βa130βa230βd3p1xp1yv1xv1yp2xp2yv2xv2yp3xp3yv3xv3y=d1a21a31a12d2a32a13a23d3[αβ]I2Xi=L[αβ]I2Xi

其中
d 1 = a 12 + a 13 + ⋯ d_1 = a_{12} + a_{13} + \cdots d1=a12+a13+,
d 2 = a 21 + a 23 + ⋯ d_2 = a_{21} + a_{23} + \cdots d2=a21+a23+,
d 3 = a 31 + a 32 + ⋯ d_3 = a_{31} + a_{32} + \cdots d3=a31+a32+,
⋯ \cdots .


3.2 静态一致性控制协议

控制协议为
u i = α ∑ j ∈ N i a i j ( p j − p i ) − β v i ( ) u_i = \red{\alpha} \sum_{j \in N_i} a_{ij} (p_j - p_i) - \blue{\beta} v_i \tag{} ui=αjNiaij(pjpi)βvi()

其中
a i j a_{ij} aij:表示邻接矩阵
α \red{\alpha} α:表示控制参数
β \blue{\beta} β:表示控制参数


通过分析控制协议:
u 1 = α a 11 ( p 1 − p 1 ) + α a 12 ( p 2 − p 1 ) + α a 13 ( p 3 − p 1 ) + ⋯ + β v 1 u 2 = α a 21 ( p 1 − p 2 ) + α a 22 ( p 2 − p 2 ) + α a 23 ( p 3 − p 2 ) + ⋯ + β v 2 u 3 = α a 31 ( p 1 − p 3 ) + α a 32 ( p 2 − p 3 ) + α a 33 ( p 3 − p 3 ) + ⋯ + β v 3 ( ) \begin{aligned} u_1 &= \red{\alpha} a_{11} (p_1 - p_1) + \red{\alpha} a_{12} (p_2 - p_1) + \red{\alpha} a_{13} (p_3 - p_1) + \cdots + \blue{\beta} v_1 \\ u_2 &= \red{\alpha} a_{21} (p_1 - p_2) + \red{\alpha} a_{22} (p_2 - p_2) + \red{\alpha} a_{23} (p_3 - p_2) + \cdots + \blue{\beta} v_2 \\ u_3 &= \red{\alpha} a_{31} (p_1 - p_3) + \red{\alpha} a_{32} (p_2 - p_3) + \red{\alpha} a_{33} (p_3 - p_3) + \cdots + \blue{\beta} v_3 \\ \end{aligned} \tag{} u1u2u3=αa11(p1p1)+αa12(p2p1)+αa13(p3p1)++βv1=αa21(p1p2)+αa22(p2p2)+αa23(p3p2)++βv2=αa31(p1p3)+αa32(p2p3)+αa33(p3p3)++βv3()

可以得到以下控制协议的矩阵形式:

[ u 1 x u 1 y u 2 x u 2 y u 3 x u 3 y ] = [ − α d 1 0 − β 0 α a 12 0 0 0 α a 13 0 0 0 0 − α d 1 0 − β 0 α a 12 0 0 0 α a 13 0 0 α a 21 0 0 0 − α d 2 0 − β 0 α a 23 0 0 0 0 α a 21 0 0 0 − α d 2 0 − β 0 α a 23 0 0 α a 31 0 0 0 α a 32 0 0 0 − α d 3 0 − β 0 0 α a 31 0 0 0 α a 32 0 0 0 − α d 3 0 − β ] [ p 1 x p 1 y v 1 x v 1 y p 2 x p 2 y v 2 x v 2 y p 3 x p 3 y v 3 x v 3 y ] = ( [ − d 1 a 12 a 13 a 21 − d 2 a 23 a 31 a 32 − d 3 ] ⊗ [ α 0 ] ⊗ I 2 + [ − 1 0 0 0 − 1 0 0 0 − 1 ] ⊗ [ 0 β ] ⊗ I 2 ) ⋅ X i = ( − L ⊗ [ α 0 ] ⊗ I 2 + − I N ⊗ [ 0 β ] ⊗ I 2 ) ⋅ X i \begin{aligned} \left[\begin{matrix} u_1^x \\ u_1^y \\ u_2^x \\ u_2^y \\ u_3^x \\ u_3^y \\ \end{matrix}\right]&= \left[\begin{matrix} -\red{\alpha}d_1 & 0 & -\blue{\beta} & 0 & \red{\alpha}a_{12} & 0 & 0 & 0 & \red{\alpha}a_{13} & 0 & 0 & 0 \\ 0 & -\red{\alpha}d_1 & 0 & -\blue{\beta} & 0 & \red{\alpha}a_{12} & 0 & 0 & 0 & \red{\alpha}a_{13} & 0 & 0 \\ \red{\alpha}a_{21} & 0 & 0 & 0 & -\red{\alpha}d_{2} & 0 & -\blue{\beta} & 0 & \red{\alpha}a_{23} & 0 & 0 & 0 \\ 0 & \red{\alpha}a_{21} & 0 & 0 & 0 & -\red{\alpha}d_{2} & 0 & -\blue{\beta} & 0 & \red{\alpha}a_{23} & 0 & 0 \\ \red{\alpha}a_{31} & 0 & 0 & 0 & \red{\alpha}a_{32} & 0 & 0 & 0 & -\red{\alpha}d_{3} & 0 & -\blue{\beta} & 0 \\ 0 & \red{\alpha}a_{31} & 0 & 0 & 0 & \red{\alpha}a_{32} & 0 & 0 & 0 & -\red{\alpha}d_{3} & 0 & -\blue{\beta} \\ \end{matrix}\right] \left[\begin{matrix} p_1^x \\ p_1^y \\ v_1^x \\ v_1^y \\ p_2^x \\ p_2^y \\ v_2^x \\ v_2^y \\ p_3^x \\ p_3^y \\ v_3^x \\ v_3^y \\ \end{matrix}\right] \\ &=( \left[\begin{matrix} -d_{1} & a_{12} & a_{13} \\ a_{21} & -d_{2} & a_{23} \\ a_{31} & a_{32} & -d_{3} \\ \end{matrix}\right] \otimes \left[\begin{matrix} \red{\alpha} & 0 \end{matrix}\right] \otimes I_2 + \left[\begin{matrix} -1 & 0 & 0 \\ 0 & -1 & 0 \\ 0 & 0 & -1 \\ \end{matrix}\right] \otimes \left[\begin{matrix} 0 & \blue{\beta} \end{matrix}\right] \otimes I_2) \cdot X_i \\ &= (\red{-L \otimes \left[\begin{matrix} \red{\alpha} & 0 \end{matrix}\right] \otimes I_2} + \red{-I_N \otimes \left[\begin{matrix} 0 & \blue{\beta} \end{matrix}\right] \otimes I_2}) \cdot X_i \end{aligned} u1xu1yu2xu2yu3xu3y=αd10αa210αa3100αd10αa210αa31β000000β0000αa120αd20αa3200αa120αd20αa3200β000000β00αa130αa230αd300αa130αa230αd30000β000000βp1xp1yv1xv1yp2xp2yv2xv2yp3xp3yv3xv3y=(d1a21a31a12d2a32a13a23d3[α0]I2+100010001[0β]I2)Xi=(L[α0]I2+IN[0β]I2)Xi

其中
d 1 = a 12 + a 13 + ⋯ d_1 = a_{12} + a_{13} + \cdots d1=a12+a13+,
d 2 = a 21 + a 23 + ⋯ d_2 = a_{21} + a_{23} + \cdots d2=a21+a23+,
d 3 = a 31 + a 32 + ⋯ d_3 = a_{31} + a_{32} + \cdots d3=a31+a32+,
⋯ \cdots ,
下标 N _N N 表示智能体的个数。

Logo

更多推荐