This article considers a consensus problem of multiagent systems with double integrator dynamics under nonuniform sampling. In the considered problem, the maximum sampling time can be selected arbitrarily. Moreover, the communication graph can change to any possible topology as long as its associated graph Laplacian has eigenvalues in an arbitrarily selected region. Existence of a controller that ensures consensus in this setting is shown when the changing topology graphs are undirected and have a spanning tree. Also, explicit bounds for controller parameters are given. A sufficient condition is given to solve the consensus problem based on making the closed loop system matrix a contraction using a particular coordinate system for general linear dynamics. It is shown that the given condition immediately generalizes to changing topology in the case of undirected topology graphs. This condition is applied to double integrator dynamics to obtain explicit bounds on the controller.
Citation: Ufuk Sevim, Leyla Goren-Sumer. Consensus of double integrator multiagent systems under nonuniform sampling and changing topology[J]. AIMS Mathematics, 2023, 8(7): 16175-16190. doi: 10.3934/math.2023827
[1] | He Yuan, Zhuo Liu . Lie n-centralizers of generalized matrix algebras. AIMS Mathematics, 2023, 8(6): 14609-14622. doi: 10.3934/math.2023747 |
[2] | Xinfeng Liang, Mengya Zhang . Triangular algebras with nonlinear higher Lie n-derivation by local actions. AIMS Mathematics, 2024, 9(2): 2549-2583. doi: 10.3934/math.2024126 |
[3] | Anas Al-Masarwah, Nadeen Kdaisat, Majdoleen Abuqamar, Kholood Alsager . Crossing cubic Lie algebras. AIMS Mathematics, 2024, 9(8): 22112-22129. doi: 10.3934/math.20241075 |
[4] | Shan Li, Kaijia Luo, Jiankui Li . Generalized Lie n-derivations on generalized matrix algebras. AIMS Mathematics, 2024, 9(10): 29386-29403. doi: 10.3934/math.20241424 |
[5] | He Yuan, Qian Zhang, Zhendi Gu . Characterizations of generalized Lie n-higher derivations on certain triangular algebras. AIMS Mathematics, 2024, 9(11): 29916-29941. doi: 10.3934/math.20241446 |
[6] | Nouf Almutiben, Ryad Ghanam, G. Thompson, Edward L. Boone . Symmetry analysis of the canonical connection on Lie groups: six-dimensional case with abelian nilradical and one-dimensional center. AIMS Mathematics, 2024, 9(6): 14504-14524. doi: 10.3934/math.2024705 |
[7] | Xianguo Hu . Universal enveloping Hom-algebras of regular Hom-Poisson algebras. AIMS Mathematics, 2022, 7(4): 5712-5727. doi: 10.3934/math.2022316 |
[8] | Nouf Almutiben, Edward L. Boone, Ryad Ghanam, G. Thompson . Classification of the symmetry Lie algebras for six-dimensional co-dimension two Abelian nilradical Lie algebras. AIMS Mathematics, 2024, 9(1): 1969-1996. doi: 10.3934/math.2024098 |
[9] | Baiying He, Siyu Gao . The nonisospectral integrable hierarchies of three generalized Lie algebras. AIMS Mathematics, 2024, 9(10): 27361-27387. doi: 10.3934/math.20241329 |
[10] | Mohd Arif Raza, Huda Eid Almehmadi . Lie (Jordan) σ−centralizer at the zero products on generalized matrix algebra. AIMS Mathematics, 2024, 9(10): 26631-26648. doi: 10.3934/math.20241295 |
This article considers a consensus problem of multiagent systems with double integrator dynamics under nonuniform sampling. In the considered problem, the maximum sampling time can be selected arbitrarily. Moreover, the communication graph can change to any possible topology as long as its associated graph Laplacian has eigenvalues in an arbitrarily selected region. Existence of a controller that ensures consensus in this setting is shown when the changing topology graphs are undirected and have a spanning tree. Also, explicit bounds for controller parameters are given. A sufficient condition is given to solve the consensus problem based on making the closed loop system matrix a contraction using a particular coordinate system for general linear dynamics. It is shown that the given condition immediately generalizes to changing topology in the case of undirected topology graphs. This condition is applied to double integrator dynamics to obtain explicit bounds on the controller.
A Poisson algebra is a triple, (L,⋅,[−,−]), where (L,⋅) is a commutative associative algebra and (L,[−,−]) is a Lie algebra that satisfies the following Leibniz rule:
[x,y⋅z]=[x,y]⋅z+y⋅[x,z],∀x,y,z∈L. |
Poisson algebras appear naturally in the study of Hamiltonian mechanics and play a significant role in mathematics and physics, such as in applications of Poisson manifolds, integral systems, algebraic geometry, quantum groups, and quantum field theory (see [7,11,24,25]). Poisson algebras can be viewed as the algebraic counterpart of Poisson manifolds. With the development of Poisson algebras, many other algebraic structures have been found, such as Jacobi algebras [1,9], Poisson bialgebras [20,23], Gerstenhaber algebras, Lie-Rinehart algebras [16,17,26], F-manifold algebras [12], Novikov-Poisson algebras [28], quasi-Poisson algebras [8] and Poisson n-Lie algebras [10].
As a dual notion of a Poisson algebra, the concept of a transposed Poisson algebra was recently introduced by Bai et al. [2]. A transposed Poisson algebra (L,⋅,[−,−]) is defined by exchanging the roles of the two binary operations in the Leibniz rule defining the Poisson algebra:
2z⋅[x,y]=[z⋅x,y]+[x,z⋅y],∀x,y,z∈L, |
where (L,⋅) is a commutative associative algebra and (L,[−,−]) is a Lie algebra.
It is shown that a transposed Poisson algebra possesses many important identities and properties and can be naturally obtained by taking the commutator in the Novikov-Poisson algebra [2]. There are many results on transposed Poisson algebras, such as those on transposed Hom-Poisson algebras [18], transposed BiHom-Poisson algebras [21], a bialgebra theory for transposed Poisson algebras [19], the relation between 12-derivations of Lie algebras and transposed Poisson algebras [14], the relation between 12-biderivations and transposed Poisson algebras [29], and the transposed Poisson structures with fixed Lie algebras (see [6] for more details).
The notion of an n-Lie algebra (see Definition 2.1), as introduced by Filippov [15], has found use in many fields in mathematics and physics [4,5,22,27]. The explicit construction of n-Lie algebras has become one of the important problems in this theory. In [3], Bai et al. gave a construction of (n+1)-Lie algebras through the use of n-Lie algebras and some linear functions. In [13], Dzhumadil′daev introduced the notion of a Poisson n-Lie algebra which can be used to construct an (n+1)-Lie algebra under an additional strong condition. In [2], Bai et al. showed that this strong condition for n=2 holds automatically for a transposed Poisson algebra, and they gave a construction of 3-Lie algebras from transposed Poisson algebras with derivations. They also found that this constructed 3-Lie algebra and the commutative associative algebra satisfy the analog of the compatibility condition for transposed Poisson algebras, which is called a transposed Poisson 3-Lie algebra. This motivated them to introduce the concept of a transposed Poisson n-Lie algebra (see Definition 2.2) and propose the following conjecture:
Conjecture 1.1. [2] Let n≥2 be an integer and (L,⋅,μn) a transposed Poisson n-Lie algebra. Let D be a derivation of (L,⋅) and (L,μn). Define an (n+1)-ary operation:
μn+1(x1,⋯,xn+1):=n+1∑i=1(−1)i−1D(xi)μn(x1,⋯,ˆxi,⋯,xn+1),∀x1,⋯,xn+1∈L, |
where ˆxi means that the i-th entry is omitted. Then, (L,⋅,μn+1) is a transposed Poisson (n+1)-Lie algebra.
In this paper, based on the identities for transposed Poisson n-Lie algebras given in Section 2, we prove that Conjecture 1.1 holds under a certain strong condition described in Section 3 (see Definition 2.3 and Theorem 3.2).
Throughout the paper, all vector spaces are taken over a field of characteristic zero. To simplify notations, the commutative associative multiplication (⋅) will be omitted unless the emphasis is needed.
In this section, we first recall some definitions, and then we exhibit a class of identities for transposed Poisson n-Lie algebras.
Definition 2.1. [15] Let n≥2 be an integer. An n-Lie algebra is a vector space L, together with a skew-symmetric linear map [−,⋯,−]:⊗nL→L, such that, for any xi,yj∈L,1≤i≤n−1,1≤j≤n, the following identity holds:
[[y1,⋯,yn],x1,⋯,xn−1]=n∑i=1(−1)i−1[[yi,x1,⋯,xn−1],y1,⋯,ˆyi,⋯,yn]. | (2.1) |
Definition 2.2. [2] Let n≥2 be an integer and L a vector space. The triple (L,⋅,[−,⋯,−]) is called a transposed Poisson n-Lie algebra if (L,⋅) is a commutative associative algebra and (L,[−,⋯,−]) is an n-Lie algebra such that, for any h,xi∈L,1≤i≤n, the following identity holds:
nh[x1,⋯,xn]=n∑i=1[x1,⋯,hxi,⋯,xn]. | (2.2) |
Some identities for transposed Poisson algebras in [2] can be extended to the following theorem for transposed Poisson n-Lie algebras.
Theorem 2.1. Let (L,⋅,[−,⋯,−]) be a transposed Poisson n-Lie algebra. Then, the following identities hold:
(1) For any xi∈L,1≤i≤n+1, we have
n+1∑i=1(−1)i−1xi[x1,⋯,ˆxi,⋯,xn+1]=0; | (2.3) |
(2) For any h,xi,yj∈L,1≤i≤n−1,1≤j≤n, we have
n∑i=1(−1)i−1[h[yi,x1,⋯,xn−1],y1,⋯,ˆyi,⋯,yn]=[h[y1,⋯,yn],x1,⋯,xn−1]; | (2.4) |
(3) For any xi,yj∈L,1≤i≤n−1,1≤j≤n+1, we have
n+1∑i=1(−1)i−1[yi,x1,⋯,xn−1][y1,⋯,ˆyi,⋯,yn+1]=0; | (2.5) |
(4) For any x1,x2,yi∈L,1≤i≤n, we have
n∑i=1n∑j=1,j≠i[y1,⋯,yix1,⋯,yjx2,⋯,yn]=n(n−1)x1x2[y1,y2,⋯,yn]. | (2.6) |
Proof. (1) By Eq (2.2), for any 1≤i≤n+1, we have
nxi[x1,⋯,xi−1,xi+1,⋯,xn+1]=∑j≠i[x1,⋯,xi−1,xi+1,⋯,xixj,⋯,xn+1]. |
Thus, we obtain
n+1∑i=1(−1)i−1nxi[x1,⋯,ˆxi,⋯,xn+1]=n+1∑i=1n+1∑j=1,j≠i(−1)i−1[x1,⋯,ˆxi,⋯,xixj,⋯,xn+1]. |
Note that, for any i>j, we have
(−1)i−1[x1,⋯,xj−1,xixj,xj+1,⋯,ˆxi,⋯,xn]+(−1)j−1[x1,⋯,ˆxj,⋯,xi−1,xjxi,xi+1,⋯,xn]=(−1)i−1+(i−j−1)[x1,⋯,xj−1,xj+1,⋯,xi−1,xixj,xi+1,⋯,xn]+(−1)j−1[x1,⋯,ˆxj,⋯,xi−1,xjxi,xi+1,⋯,xn]=((−1)−j−2+(−1)j−1)[x1,⋯,xj−1,xj+1,⋯,xi−1,xixj,xi+1,⋯,xn]=0, |
which gives n+1∑i=1n+1∑j=1,j≠i(−1)i−1[x1,⋯,ˆxi,⋯,xixj,⋯,xn+1]=0.
Hence, we get
n+1∑i=1(−1)i−1nxi[x1,⋯,ˆxi,⋯,xn+1]=0. |
(2) By Eq (2.2), we have
−[h[y1,⋯,yn],x1,⋯,xn−1]−n−1∑i=1[[y1,⋯,yn],x1,⋯,hxi,⋯,xn−1]=−nh[[y1,⋯,yn],x1,⋯,xn−1], |
and, for any 1≤j≤n,
(−1)j−1([h[yj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn−1]+n∑i=1,i≠j[[yj,x1,⋯,xn−1],y1,⋯,hyi,⋯,ˆyj,⋯,yn−1])=(−1)j−1nh[[yj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn−1]. |
By taking the sum of the above n+1 identities and applying Eq (2.1), we get
−[h[y1,⋯,yn],x1,⋯,xn−1]−n−1∑i=1[[y1,⋯,yn],x1,⋯,hxi,⋯,xn−1]+n∑j=1(−1)j−1([h[yj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn−1]+n∑i=1,i≠j[[yj,x1,⋯,xn−1],y1,⋯,hyi,⋯,ˆyj,⋯,yn−1])=−nh[[y1,⋯,yn],x1,⋯,xn−1]+nhn∑j=1(−1)j−1[[yj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn−1]=0. |
We denote
Aj:=n∑i=1,i≠j(−1)i−1[[yi,x1,⋯,xn−1],y1,⋯,hyj,⋯,ˆyi,⋯,yn],1≤j≤n,Bi:=[[y1,⋯,yn],x1,⋯,hxi,⋯,xn−1],1≤i≤n−1. |
Then, the above equation can be rewritten as
n∑i=1(−1)i−1[h[yi,x1,⋯,xn−1],y1,⋯,ˆyi,⋯,yn]−[h[y1,⋯,yn],x1,⋯,xn−1]+n∑j=1Aj−n−1∑i=1Bi=0. | (2.7) |
By applying Eq (2.1) to Aj,1≤j≤n, we have
Aj=n∑i=1,i≠j(−1)i−1[[yi,x1,⋯,xn−1],y1,⋯,hyj,⋯,ˆyi,⋯,yn]=[[y1,⋯,hyj,⋯,yn],x1,⋯,xn−1]+(−1)j[[hyj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn]. |
Thus, we get
n∑j=1Aj=n∑j=1[[y1,⋯,hyj,⋯,yn],x1,⋯,xn−1]+n∑j=1(−1)j[[hyj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn]=n[h[y1,⋯,yn],x1,⋯,xn−1]+n∑j=1(−1)j[[hyj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn]. |
By applying Eq (2.1) to Bi,1≤i≤n−1, we have
Bi=[[y1,⋯,yn],x1,⋯,hxi,⋯,xn−1]=n∑j=1(−1)j−1[[yj,x1,⋯,hxi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]. |
Thus, we get
n−1∑i=1Bi=n−1∑i=1n∑j=1(−1)j−1[[yj,x1,⋯,hxi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]=n∑j=1n−1∑i=1(−1)j−1[[yj,x1,⋯,hxi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]. |
Note that, by Eq (2.2), we have
n−1∑i=1(−1)j−1[[yj,x1,⋯,hxi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]=(−1)j−1n[h[yj,x1,⋯,xi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]+(−1)j[[hyj,x1,⋯,xi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]. |
Thus, we obtain
n−1∑i=1Bi=n∑j=1(−1)j−1n[h[yj,x1,⋯,xi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]+n∑j=1(−1)j[[hyj,x1,⋯,xi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]. |
By substituting these equations into Eq (2.7), we have
n∑i=1(−1)i−1[h[yi,x1,⋯,xn−1],y1,⋯,ˆyi,⋯,yn]−[h[y1,⋯,yn],x1,⋯,xn−1]+n[h[y1,⋯,yn],x1,⋯,xn−1]+n∑j=1(−1)j[[hyj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,yn]−n∑j=1(−1)j−1n[h[yj,x1,⋯,xi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]−n∑j=1(−1)j[[hyj,x1,⋯,xi,⋯,xn−1],y1,⋯ˆyj,⋯,yn]=0, |
which implies that
(n−1)(n∑i=1(−1)i[h[yi,x1,⋯,xn−1],y1,⋯,ˆyi,⋯,yn]+[h[y1,⋯,yn],x1,⋯,xn−1])=0. |
Therefore, the proof of Eq (2.4) is completed.
(3) By Eq (2.2), for any 1≤j≤n+1, we have
(−1)j−1n[yj,x1,⋯,xn−1][y1,⋯,ˆyj,⋯,yn+1]=n+1∑i=1,i≠j(−1)j−1[y1,⋯,yi[yj,x1,⋯,xn−1],⋯,ˆyj,⋯,yn+1]. |
By taking the sum of the above n+1 identities, we obtain
n+1∑j=1(−1)j−1n[yj,x1,⋯,xn−1][y1,⋯,ˆyj,⋯,yn+1]=n+1∑j=1n+1∑i=1,i≠j(−1)j−1[y1,⋯,yi[yj,x1,⋯,xn−1],⋯,ˆyj,⋯,yn+1]. |
Thus, we only need to prove the following equation:
n+1∑j=1n+1∑i=1,i≠j(−1)j−1[y1,⋯,yi[yj,x1,⋯,xn−1],⋯,ˆyj,⋯,yn+1]=0. |
Note that
n+1∑j=1n+1∑i=1,i≠j(−1)j−1[y1,⋯,yi[yj,x1,⋯,xn−1],⋯,ˆyj,⋯,yn+1]=n+1∑i=1n+1∑j=1,j≠i(−1)j−1[y1,⋯,yi[yj,x1,⋯,xn−1],⋯,ˆyj,⋯,yn+1]=n+1∑i=1i−1∑j=1(−1)i+j−1[yi[yj,x1,⋯,xn−1],y1,⋯,ˆyj,⋯,ˆyi,⋯,yn+1]+n+1∑i=1n+1∑j=i+1(−1)i+j[yi[yj,x1,⋯,xn−1],y1,⋯,ˆyi,⋯,ˆyj,⋯,yn+1](2.4)=n+1∑i=1(−1)i[yi[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,xn−1](2.3)=0. |
Hence, the conclusion holds.
(4) By applying Eq (2.2), we have
n2x1x2[y1,y2,⋯,yn]=nx1n∑j=1[y1,⋯,yjx2,⋯,yn]=n∑i=1n∑j=1,j≠i[y1,⋯,yix1,⋯,yjx2,⋯,yn]+n∑j=1[y1,⋯,yjx1x2,⋯,yn]=n∑i=1n∑j=1,j≠i[y1,⋯,yix1,⋯,yjx2,⋯,yn]+nx1x2[y1,⋯,yn], |
which gives
n(n−1)x1x2[y1,y2,⋯,yn]=n∑i=1n∑j=1,j≠i[y1,⋯,yix1,⋯,yjx2,⋯,yn]. |
Hence, the proof is completed.
To prove Conjecture 1.1, we need the following extra condition.
Definition 2.3. A transposed Poisson n-Lie algebra (L,⋅,[−,⋯,−]) is called strong if the following identity holds:
y1[hy2,x1,⋯,xn−1]−y2[hy1,x1,⋯,xn−1]+n−1∑i=1(−1)i−1hxi[y1,y2,x1,⋯,ˆxi,⋯,xn−1]=0 | (2.8) |
for any y1,y2,xi∈L,1≤i≤n−1.
Remark 2.1. When n=2, the identity is
y1[hy2,x1]+y2[x1,hy1]+hx1[y1,y2]=0, |
which is exactly Theorem 2.5 (11) in [2]. Thus, in the case of a transposed Poisson algebra, the strong condition always holds. So far, we cannot prove that the strong condition fails to hold for n≥3.
Proposition 2.1. Let (L,⋅,[−,⋯,−]) be a strong transposed Poisson n-Lie algebra. Then,
y1[hy2,x1,⋯,xn−1]−hy1[y2,x1,⋯,xn−1]=y2[hy1,x1,⋯,xn−1]−hy2[y1,x1,⋯,xn−1] | (2.9) |
for any y1,y2,xi∈L,1≤i≤n−1.
Proof. By Eq (2.3), we have
−hy1[y2,x1,⋯,xn−1]+hy2[y1,x1,⋯,xn−1]=n−1∑i=1(−1)i−1hxi[y1,y2,x1,⋯,ˆxi,⋯,xn−1]. |
Then, the statement follows from Eq (2.8).
In this section, we will prove Conjecture 1.1 for strong transposed Poisson n-Lie algebras. First, we recall the notion of derivations of transposed Poisson n-Lie algebras.
Definition 3.1. Let (L,⋅,[−,⋯,−]) be a transposed Poisson n-Lie algebra. The linear operation D:L→L is called a derivation of (L,⋅,[−,⋯,−]) if the following holds for any u,v,xi∈L,1≤i≤n:
(1) D is a derivation of (L,⋅), i.e., D(uv)=D(u)v+uD(v);
(2) D is a derivation of (L,[−,⋯,−]), i.e.,
D([x1,⋯,xn])=n∑i=1[x1,⋯,xi−1,D(xi),xi+1,⋯,xn]. |
Lemma 3.1. Let (L,⋅,[−,⋯,−]) be a transposed Poisson n-Lie algebra and D a derivation of (L,⋅,[−,⋯,−]). For any yi∈L,1≤i≤n+1, we have the following:
(1)
n+1∑i=1(−1)i−1D(yi)D([y1,⋯,ˆyi,⋯,yn+1])=n+1∑i=1n+1∑j=1,j≠i(−1)i−1D(yi)[y1,⋯,D(yj),⋯,ˆyi,⋯,yn+1]; | (3.1) |
(2)
n+1∑i=1(−1)i−1D(yi)D([y1,⋯,ˆyi,⋯,yn+1])=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)iyi[y1,⋯,D(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1], | (3.2) |
where, for any i>j, j∑i denotes the empty sum, which is equal to zero.
Proof. (1) The statement follows immediately from Definition 3.1.
(2) By applying Eq (3.1), we need to prove the following equation:
n+1∑i=1n+1∑j=1,j≠i(−1)i−1nD(yi)[y1,⋯,D(yj),⋯,ˆyi,⋯,yn+1]=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)inyi[y1,⋯,D(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1]. |
For any 1≤i≤n+1, denote Ai:=nn+1∑j=1,j≠i(−1)i−1D(yi)[y1,⋯,D(yj),⋯,ˆyi,⋯,yn+1]. Then, we have
n+1∑i=1n+1∑j=1,j≠i(−1)i−1nD(yi)[y1,⋯,D(yj),⋯,ˆyi,⋯,yn+1]=n+1∑i=1Ai. |
Note that
Ai=(−1)i−1(nD(yi)[D(y1),y2,⋯,ˆyi,⋯,yn+1]+nD(yi)[y1,D(y2),y3,⋯,ˆyi,⋯,yn+1]+⋯+nD(yi)[y1,⋯,ˆyi,⋯,yn,D(yn+1)])=(−1)i−1([D(yi)D(y1),y2,⋯,ˆyi,⋯,yn+1]+n+1∑k=2,k≠i[D(y1),y2,⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]+[y1,D(yi)D(y2),y3,⋯,ˆyi,⋯,yn+1]+n+1∑k=1,k≠2,i[y1,D(y2),y3,⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]+⋯+[y1,⋯,ˆyi,⋯,yn,D(yi)D(yn+1)]+n∑k=1,k≠i[y1,⋯,ykD(yi),⋯,ˆyi,⋯,yn,D(yn+1)])=(−1)i−1n+1∑j=1,j≠i[y1,⋯,D(yi)D(yj),⋯,ˆyi,⋯,yn+1]+(−1)i−1n+1∑j=1,j≠in+1∑k=1,k≠j,i[y1,⋯,D(yj),⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]. |
Thus, we have
n+1∑i=1Ai=n+1∑j=1n+1∑i=1,i≠j(−1)j−1[y1,⋯,D(yj)D(yi),⋯,ˆyj,⋯,yn+1]+n+1∑i=1n+1∑j=1,j≠in+1∑k=1,k≠i,j(−1)i−1[y1,⋯,D(yj),⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]=T1+T2, |
where
T1:=n+1∑j=1n+1∑i=1,i≠j(−1)j−1[y1,⋯,D(yj)D(yi),⋯,ˆyj,⋯,yn+1],T2:=n+1∑i=1n+1∑j=1,j≠in+1∑k=1,k≠i,j(−1)i−1[y1,⋯,D(yj),⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]. |
Note that
T1=n+1∑j,i=1Bji, |
where Bji=(−1)j−1[y1,⋯,D(yj)D(yi),⋯,ˆyj,⋯,yn+1] for any 1≤j≠i≤n+1, and Bii=0 for any 1≤i≤n+1.
For any 1≤i,j≤n+1, without loss of generality, assume that i<j; then, we have
Bji+Bij=(−1)j−1[y1,⋯,D(yj)D(yi),⋯,ˆyj,⋯,yn+1]+(−1)i−1[y1,⋯,ˆyi,⋯,D(yi)D(yj),⋯,yn+1]=(−1)j−1[y1,⋯,D(yj)D(yi),⋯,ˆyj,⋯,yn+1]+(−1)i−1+j−i+1[y1,⋯,D(yj)D(yi),⋯,ˆyj,⋯,yn+1]=0, |
which implies that T1=n+1∑j,i=1Bji=0.
Thus, we get that n+1∑i=1Ai=T2.
We rewrite
n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)inyi[y1,⋯,D(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1]=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠in+1∑t=1,t≠j,k,i(−1)i⋅[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1]+n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,yiD(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1]+n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=M1+M2+M3, |
where
M1:=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠in+1∑t=1,t≠j,k,i(−1)i⋅[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1],M2:=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,yiD(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1],M3:=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]. |
Note that
M1=n+1∑i=1n+1∑j=1,j≠in+1∑k=j+1,k≠in+1∑t=1,t≠j,k,i(−1)i⋅[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1]=n+1∑i,j,k,t=1Bijkt, |
where
Bijkt={0,if any two indices are equal or k<j;(−1)i[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1],otherwise. |
For any 1≤j,k≤n+1, without loss of generality, assume that t<i; then, we have
Bijkt+Btjki=(−1)i[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1]+(−1)t[y1,⋯,D(yj),⋯,D(yk),⋯,ˆyt,⋯,ytyi,⋯,yn+1]=(−1)i[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1]+(−1)t+i−t−1[y1,⋯,D(yj),⋯,D(yk),⋯,ytyi,⋯,ˆyi,⋯,yn+1]=0, |
which implies that M1=0.
Therefore, we only need to prove the following equation:
M2+M3=n+1∑i=1n+1∑j=1,j≠in+1∑k=1,k≠i,j(−1)i−1[y1,⋯,D(yj),⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]. |
First, we have
n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,yiD(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1]+n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=n+1∑k=1,k≠ik−1∑j=1,j≠i(−1)i[y1,⋯,yiD(yj),⋯,D(yk),⋯,ˆyi,⋯,yn+1]+n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=n+1∑j=1,j≠ij−1∑k=1,k≠i(−1)i[y1,⋯,yiD(yk),⋯,D(yj),⋯,ˆyi,⋯,yn+1]+n+1∑j=1,j≠in+1∑k=j+1,k≠i(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=n+1∑j=1,j≠in+1∑k=1,k≠i,j(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]. |
Thus,
M2+M3=n+1∑i=1n+1∑j=1,j≠in+1∑k=1,k≠i,j(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=n+1∑j=1n+1∑i=1,i≠jn+1∑k=1,k≠i,j(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=n+1∑j=1n+1∑i=1,i≠ji−1∑k=1,k≠j(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]+n+1∑j=1n+1∑i=1,i≠jn+1∑k=i+1,k≠j(−1)i[y1,⋯,D(yj),⋯,ˆyi,⋯,yiD(yk),⋯,yn+1]. |
Note that, for any 1≤j≤n+1, we have
n+1∑i=1,i≠ji−1∑k=1,k≠j(−1)i[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyi,⋯,yn+1]=n+1∑i=1,i≠ji−1∑k=1,k≠j(−1)i[y1,⋯,D(yj),⋯,yk−1,yiD(yk),yk+1,⋯,ˆyi⋯,yn+1]=n+1∑i=1,i≠ji−1∑k=1,k≠j(−1)k−1[y1,⋯,D(yj),⋯,ˆyk,⋯,yi−1,yiD(yk),yi+1,⋯,yn+1]=n+1∑i=1,i≠ji−1∑k=1,k≠j(−1)k−1[y1,⋯,D(yj),⋯,ˆyk,⋯,yiD(yk),⋯,yn+1]. |
Similarly, we have
n+1∑i=1,i≠jn+1∑k=i+1,k≠j(−1)i[y1,⋯,D(yj),⋯,ˆyi,⋯,yiD(yk),⋯,yn+1]=n+1∑i=1,i≠jn+1∑k=i+1,k≠j(−1)k−1[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyk,⋯,yn+1]. |
Thus,
M2+M3=n+1∑j=1n+1∑i=1,i≠ji−1∑k=1,k≠j(−1)k−1[y1,⋯,D(yj),⋯,ˆyk,⋯,yiD(yk),⋯,yn+1]+n+1∑j=1n+1∑i=1,i≠jn+1∑k=i+1,k≠j(−1)k−1[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyk,⋯,yn+1]=n+1∑j=1n+1∑i=1,i≠jn+1∑k=1,k≠i,j(−1)k−1[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyk,⋯,yn+1]=n+1∑k=1n+1∑j=1,j≠kn+1∑i=1,i≠j,k(−1)k−1[y1,⋯,D(yj),⋯,yiD(yk),⋯,ˆyk,⋯,yn+1]=n+1∑i=1n+1∑j=1,j≠in+1∑k=1,k≠j,i(−1)i−1[y1,⋯,D(yj),⋯,ykD(yi),⋯,ˆyi,⋯,yn+1]. |
The proof is completed.
Theorem 3.1. Let (L,⋅,[−,⋯,−]) be a strong transposed Poisson n-Lie algebra and D a derivation of (L,⋅,[−,⋯,−]). Define an (n+1)-ary operation:
μn+1(x1,⋯,xn+1):=n+1∑i=1(−1)i−1D(xi)[x1,⋯,ˆxi,⋯,xn+1] | (3.3) |
for any xi∈L,1≤i≤n+1. Then, (L,μn+1) is an (n+1)-Lie algebra.
Proof. For convenience, we denote
μn+1(x1,⋯,xn+1):=[x1,⋯,xn+1]. |
On one hand, we have
[[y1,⋯,yn+1],x1,⋯,xn](3.3)=n+1∑i=1(−1)i−1[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,xn](3.3)=n+1∑i=1(−1)i−1D(D(yi)[y1,⋯,ˆyi,⋯,yn+1])[x1,⋯,xn]+n+1∑i=1n∑j=1(−1)i+j−1D(xj)[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]=n+1∑i=1(−1)i−1D2(yi)[y1,⋯,ˆyi,⋯,yn+1][x1,⋯,xn]+n+1∑i=1(−1)i−1D(yi)D([y1,⋯,ˆyi,⋯,yn+1])[x1,⋯,xn]+n+1∑i=1n∑j=1(−1)i+j−1D(xj)[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn](3.1)=n+1∑i=1(−1)i−1D2(yi)[y1,⋯,ˆyi,⋯,yn+1][x1,⋯,xn]+n+1∑i=1n+1∑k=1,k≠i(−1)i−1D(yi)[y1,⋯,D(yk),⋯,ˆyi,⋯,yn+1][x1,⋯,xn]+n+1∑i=1n∑j=1(−1)i+j−1D(xj)[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]=n+1∑i=1(−1)i−1D2(yi)[y1,⋯,ˆyi,⋯,yn+1][x1,⋯,xn]+n+1∑k=1k−1∑i=1(−1)k+i−1D(yi)[D(yk),y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1][x1,⋯,xn]+n+1∑k=1n+1∑i=k+1(−1)i+kD(yi)[D(yk),y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1][x1,⋯,xn]+n+1∑i=1n∑j=1(−1)i+j−1D(xj)[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]. |
On the other hand, for any 1≤k≤n, we have
(−1)k−1[[yk,x1,⋯,xn],y1,⋯,ˆyk,⋯,yn+1](3.3)=(−1)k−1[D(yk)[x1,⋯,xn],y1,⋯,ˆyk,⋯,yn+1]+n∑j=1(−1)j+k−1[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyk,⋯,yn+1](3.3)=(−1)k−1D(D(yk)[x1,⋯,xn])[y1,⋯,ˆyk,⋯,yn+1]+k−1∑i=1(−1)i+k−1D(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]+n+1∑i=k+1(−1)i+kD(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1]+n∑j=1(−1)j+k−1D(D(xj)[yk,x1,⋯,ˆxj,⋯,xn])[y1,⋯,ˆyk,⋯,yn+1]+n∑j=1n+1∑i=k+1((−1)i+jD(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1])+n∑j=1k−1∑i=1((−1)i+j−1D(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1])=(−1)k−1D2(yk)[x1,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+k−1∑i=1(−1)i+k−1D(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]+n+1∑i=k+1(−1)i+kD(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1]+n∑j=1(−1)j+k−1D2(xj)[yk,x1,⋯,ˆxj,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+n∑j=1n+1∑i=k+1((−1)i+jD(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1])+n∑j=1k−1∑i=1((−1)i+j−1D(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1])+(−1)k−1D(yk)D([x1,⋯,xn])[y1,⋯,ˆyk,⋯,yn+1]+n∑j=1(−1)j+k−1D(xj)D([yk,x1,⋯,ˆxj,⋯,xn])[y1,⋯,ˆyk,⋯,yn+1](3.2)=(−1)k−1D2(yk)[x1,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+k−1∑i=1(−1)i+k−1D(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]+n+1∑i=k+1(−1)i+kD(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1]+n∑j=1(−1)j+k−1D2(xj)[yk,x1,⋯,ˆxj,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+n∑j=1n+1∑i=k+1((−1)i+jD(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1])+n∑j=1k−1∑i=1((−1)i+j−1D(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1])+n∑j=1n∑t=j+1(−1)kyk[x1,⋯,D(xj),⋯,D(xt),⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+n∑i=1n∑j=1,j≠i(−1)k+ixi[D(yk),x1,⋯,D(xj),⋯,ˆxi,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+n∑i=1n∑j=1,j≠in∑t=j+1,t≠i(−1)k+ixi[yk,x1,⋯,D(xj),D(xt),⋯,ˆxi,⋯,xn]⋅[y1,⋯,ˆyk,⋯,yn+1]. |
We denote
n+1∑i=1(−1)i−1[[yi,x1,⋯,xn],y1,⋯,ˆyi,⋯,yn+1]=7∑i=1Ai, |
where
A1:=n+1∑i=1(−1)i−1D2(yi)[x1,⋯,xn][y1,⋯,ˆyi,⋯,yn+1],A2:=n+1∑k=1n∑j=1(−1)k+j−1D2(xj)[yk,x1,⋯,ˆxj,⋯,xn][y1,⋯,ˆyk,⋯,yn+1],A3:=n+1∑i=1n∑j=1n∑k=j+1(−1)iyi[x1,⋯,D(xj),⋯,D(xk),⋯,xn][y1,⋯,ˆyi,⋯,yn+1],A4:=n+1∑k=1n∑i=1n∑j=1,j≠in∑t=j+1,t≠i((−1)k+ixi[yk,x1,⋯,D(xj),D(xt),⋯,ˆxi,⋯,xn]⋅[y1,⋯,ˆyk,⋯,yn+1]),A5:=n+1∑k=1k−1∑i=1(−1)k+i−1D(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]+n+1∑k=1n+1∑i=k+1(−1)i+kD(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1],A6:=n+1∑k=1n∑i=1n∑j=1,j≠i((−1)k+ixi[D(yk),x1,⋯,D(xj),⋯,ˆxi,⋯,xn]⋅[y1,⋯,ˆyk,⋯,yn+1]),A7:=n+1∑k=1n∑j=1n+1∑i=k+1((−1)k+i+jD(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1])+n+1∑k=1n∑j=1k−1∑i=1((−1)k+i+j−1D(yi)⋅[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]). |
By Eq (2.5), for fixed j, we have
n+1∑k=1(−1)k+j−1D2(xj)[yk,x1,⋯,ˆxj,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]=0. |
So, we obtain that A2=0.
By Eq (2.3), for fixed j and k, we have
n+1∑i=1(−1)iyi[x1,⋯,D(xj),⋯,D(xk),⋯,xn][y1,⋯,ˆyi,⋯,yn+1]=0. |
So, we obtain that A3=0.
By Eq (2.5), for fixed j and t, we have
n+1∑k=1(−1)k+ixi[yk,x1,⋯,D(xj),D(xt),⋯,ˆxi,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]=0. |
So, we obtain that A4=0.
By Eq (2.9), for fixed i and k, we have
(−1)k+i−1D(yi)[D(yk)[x1,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]+(−1)i+kD(yk)[D(yi)[x1,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]=(−1)k+i−1D(yi)[D(yk),y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1][x1,⋯,xn]+(−1)i+kD(yk)[D(yi),y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1][x1,⋯,xn]. |
Thus, we obtain
A5=n+1∑k=1k−1∑i=1(−1)k+i−1D(yi)[D(yk),y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1][x1,⋯,xn]+n+1∑k=1n+1∑i=k+1(−1)i+kD(yi)[D(yk),y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1][x1,⋯,xn]. |
By Eq (2.3), for fixed j and k, we have
n∑i=1(−1)k+ixi[D(yk),x1,⋯,D(xj),⋯,ˆxi,⋯,xn]=(−1)k−1D(yk)[x1,⋯,D(xj),⋯,xn]+(−1)k+j−1D(xj)[D(yk),x1,⋯,xn]=(−1)k+jD(yk)[D(xj),x1,⋯,ˆxj,⋯,xn]+(−1)k+j−1D(xj)[D(yk),x1,⋯,xn]. |
Thus, we get
A6=n+1∑k=1n∑j=1(−1)k+jD(yk)[D(xj),x1,⋯,ˆxj,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]+n+1∑k=1n∑j=1(−1)k+j−1D(xj)[D(yk),x1,⋯,xn][y1,⋯,ˆyk,⋯,yn+1]. |
By Eq (2.4), for fixed j and i, we have
n+1∑k=i+1(−1)k+i+j−1D(yi)[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyi,⋯,ˆyk,⋯,yn+1]+i−1∑k=1(−1)k+i+jD(yi)[D(xj)[yk,x1,⋯,ˆxj,⋯,xn],y1,⋯,ˆyk,⋯,ˆyi,⋯,yn+1]=(−1)j+i−1D(yi)[D(xj)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]. |
So, we obtain
A7=n∑j=1n+1∑i=1(−1)j+i−1D(yi)[D(xj)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]. |
By Eq (2.9), we have
(−1)i+jD(yi)[D(xj),x1,⋯,ˆxj,⋯,xn][y1,⋯,ˆyi,⋯,yn+1]+(−1)i+j−1D(xj)[D(yi),x1,⋯,xn][y1,⋯,ˆyi,⋯,yn+1]+(−1)j+i−1D(yi)[D(xj)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]=(−1)j+i−1D(xj)[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]. |
So, we get
A6+A7=n+1∑i=1n∑j=1(−1)j+i−1D(xj)[D(yi)[y1,⋯,ˆyi,⋯,yn+1],x1,⋯,ˆxj,⋯,xn]. |
Thus, we have
7∑i=1Ai=A1+A5+A6+A7=[[y1,⋯,yn+1],x1,⋯,xn]. |
Therefore, (L,μn+1) is an (n+1)-Lie algebra.
Now, we can prove Conjecture 1.1 for strong transposed Poisson n-Lie algebras.
Theorem 3.2. With the notations in Theorem 3.1, (L,⋅,μn+1) is a strong transposed Poisson (n+1)-Lie algebra.
Proof. For convenience, we denote μn+1(x1,⋯,xn+1):=[x1,⋯,xn+1]. According to Theorem 3.1, we only need to prove Eqs (2.2) and (2.8).
Proof of Eq (2.2). By Eq (3.3), we have
n+1∑i=1[x1,⋯,hxi,⋯,xn+1]=D(hx1)[x2,⋯,xn+1]+n+1∑j=2(−1)j−1D(xj)[hx1,x2,⋯,ˆxj,⋯,xn+1]−D(hx2)[x1,x3,⋯,xn+1]+n+1∑j=1,j≠2(−1)j−1D(xj)[x1,hx2,x3,⋯,ˆxj,⋯,xn+1]+⋯+(−1)nD(hxn)[x1,⋯,xn]+n∑j=1(−1)j−1D(xj)[x1,⋯,ˆxj,⋯,xn,hxn+1]=n+1∑i=1(−1)i−1D(hxi)[x1,⋯,ˆxi,⋯,xn]+n+1∑i=1n+1∑j=1,j≠i(−1)j−1D(xj)[x1,,⋯,hxi,⋯,ˆxj,⋯,xn+1]=n+1∑i=1(−1)i−1D(hxi)[x1,⋯,ˆxi,⋯,xn+1]+n+1∑j=1n+1∑i=1,i≠j(−1)j−1D(xj)[x1,⋯,hxi,⋯,ˆxj,⋯,xn+1]=n+1∑i=1(−1)i−1hD(xi)[x1,⋯,ˆxi,⋯,xn+1]+n+1∑i=1(−1)i−1xiD(h)[x1,⋯,ˆxi,⋯,xn+1]+n+1∑j=1n+1∑i=1,i≠j(−1)j−1D(xj)[x1,⋯,hxi,⋯,ˆxj,⋯,xn+1](2.3)=n+1∑i=1(−1)i−1hD(xi)[x1,⋯,ˆxi,⋯,xn+1]+n+1∑j=1n+1∑i=1,i≠j(−1)j−1D(xj)[x1,⋯,hxi,⋯,ˆxj,⋯,xn+1](3.3)=h[x1,⋯,xn+1]+n+1∑j=1n+1∑i=1,i≠j(−1)j−1D(xj)[x1,⋯,hxi,⋯,ˆxj,⋯,xn+1](2.2)=h[x1,⋯,xn+1]+nhn+1∑j=1(−1)j−1D(xj)[x1,⋯,ˆxj,⋯,xn+1](3.3)=h[x1,⋯,xn+1]+nh[x1,⋯,xn+1]=(n+1)h[x1,⋯,xn+1]. |
Proof of Eq (2.8). By Eq (3.3), we have
y1[hy2,x1,⋯,xn]−y2[hy1,x1,⋯,xn]+n∑i=1(−1)i−1hxi[y1,y2,x1,⋯,ˆxi,⋯,xn]=y1y2D(h)[x1,⋯,xn]+y1hD(y2)[x1,⋯,xn]−y1D(x1)[hy2,x2,⋯,xn]+y1D(x2)[hy2,x1,x3,⋯,xn]+⋯+(−1)ny1D(xn)[hy2,x1,⋯,xn−1]−y2y1D(h)[x1,⋯,xn]−y2hD(y1)[x1,⋯,xn]+y2D(x1)[hy1,x2,⋯,xn]−y2D(x2)[hy1,x1,x3,⋯,xn]+⋯+(−1)n−1y2D(xn)[hy1,x1,⋯,xn−1]+hx1D(y1)[y2,x2,⋯,xn]−hx1D(y2)[y1,x2,⋯,xn]+hx1D(x2)[y1,y2,x3,⋯,xn]+⋯+(−1)n+1hx1D(xn)[y1,y2,x2,⋯,xn−1]−hx2D(y1)[y2,x1,x3,⋯,xn]+hx2D(y2)[y1,x1,x3,⋯,xn]−hx2D(x1)[y1,y2,x3,⋯,xn]+⋯+(−1)n+2hx2D(xn)[y1,y2,x1,x3,⋯,xn−1]+⋯+(−1)n−1hxnD(y1)[y2,x1,⋯,xn−1]+(−1)nhxnD(y2)[y1,x1,⋯,xn−1]+(−1)n+1hxnD(x1)[y1,y2,x2,⋯,xn−1]+⋯+(−1)2n−1hxnD(xn−1)[y1,y2,x1,⋯,xn−2]=−y2hD(y1)[x1,⋯,xn]+hx1D(y1)[y2,x2,⋯,xn]+n∑i=2(−1)i−1hxiD(y1)[y2,x1,⋯,ˆxi,⋯,xn]+y1hD(y2)[x1,⋯,xn]−hx1D(y2)[y1,x2,⋯,xn]+n∑i=2(−1)ihxiD(y2)[y1,x1,⋯,ˆxi,⋯,xn]−y1D(x1)[hy2,x2,⋯,xn]+y2D(x1)[hy1,x2,⋯,xn]+n∑i=2(−1)i−1hxiD(x1)[y1,y2,x2,⋯,ˆxi,⋯,xn]+y1D(x2)[hy2,x1,x3,⋯,xn]−y2D(x2)[hy1,x1,x3,⋯,xn]+hx1D(x2)[y1,y2,x3,⋯,xn]+n∑i=3(−1)ihxiD(x2)[y1,y2,x1,x3,⋯,ˆxi,⋯,xn]⋯+(−1)ny1D(xn)[hy2,x1,⋯,xn−1]+(−1)n−1y2D(xn)[hy1,x1,⋯,xn−1]+n−1∑j=1(−1)n+j−1hxjD(xn)[y1,y2,x1,⋯,ˆxj,⋯,xn−1]=A1+A2+n∑i=1Bi, |
where
A1:=−y2hD(y1)[x1,⋯,xn]+n∑i=1(−1)i−1hxiD(y1)[y2,x1,⋯,ˆxi,⋯,xn],A2:=y1hD(y2)[x1,⋯,xn]+n∑i=1(−1)ihxiD(y2)[y1,x1,⋯,ˆxi,⋯,xn], |
and, for any 1≤i≤n,
Bi:=(−1)iy1D(xi)[hy2,x1,⋯,ˆxi,⋯,xn]+(−1)i−1y2D(xi)[hy1,x1,⋯,ˆxi,⋯,xn]+i−1∑j=1(−1)i+j−1hxjD(xi)[y1,y2,x1,⋯,ˆxj,⋯,ˆxi,⋯,xn]+n∑j=i+1(−1)i+jhxjD(xi)[y1,y2,x1,⋯,ˆxi,⋯,ˆxj,⋯,xn]. |
By Eq (2.3), we have
A1=hD(y1)(−y2[x1,⋯,xn]+n∑i=1(−1)i−1xi[y2,x1,⋯,ˆxi,⋯,xn])=0. |
Similarly, we have that A2=0.
By Eq (2.8), for any 1≤i≤n, we have
Bi=(−1)iD(xi)(y1[hy2,x1,⋯,ˆxi,⋯,xn]−y2[hy1,x1,⋯,ˆxi,⋯,xn]+i−1∑j=1(−1)j−1hxj[y1,y2,x1,⋯,ˆxj,⋯,ˆxi,⋯,xn]+n∑j=i+1(−1)jhxj[y1,y2,x1,⋯,ˆxi,⋯,ˆxj,⋯,xn])=0. |
Thus, we get
y1[hy2,x1,⋯,xn]−y2[hy1,x1,⋯,xn]+n∑i=1(−1)i−1hxi[y1,y2,x1,⋯,ˆxi,⋯,xn]=0. |
The proof is completed.
Example 3.1. The commutative associative algebra L=k[x1,x2,x3], together with the bracket
[x,y]:=x⋅D1(y)−y⋅D1(x),∀x,y∈L. |
gives a transposed Poisson algebra (L,⋅,[−,−]), where D1=∂x1 ([2, Proposition 2.2]). Note that the transposed Poisson algebra (L,⋅,[−,−]) is strong according to Remark 2.5. Now, let D2=∂x2; one can check that D2 is a derivation of (L,⋅,[−,−]). Then, there exists a strong transposed Poisson 3-Lie algebra defined by
[x,y,z]:=D2(x)(yD1(z)−zD1(y))+D2(y)(zD1(x)−xD1(z))+D2(z)(xD1(y)−yD1(x)), ∀x,y,z∈L. |
We note that [x1,x2,x3]=x3, which is non-zero. The strong condition can be checked as follows:
For any h,y1,y2,z1,z2∈L, by a direct calculation, we have
y1[hy2,z1,z2]=y1z1hD1(z2)D2(y2)−y1z2hD1(z1)D2(y2)+y1y2z1D1(z2)D2(h)−y1y2z2D1(z1)D2(h)−y1y2hD1(z2)D2(z1)+y1z2hD1(y2)D2(z1)+y1y2z2D1(h)D2(z1)+y1y2hD1(z1)D2(z2)−y1z1hD1(y2)D2(z2)−y1y2z1D1(h)D2(z2), |
−y2[hy1,z1,z2]=−y2z1hD1(z2)D2(y1)+y2z2hD1(z1)D2(y1)−y1y2z1D1(z2)D2(h)+y1y2z2D1(z1)D2(h)+y1y2hD1(z2)D2(z1)−y2z2hD1(y1)D2(z1)−y1y2z2D1(h)D2(z1)−y1y2hD1(z1)D2(z2)+y2z1hD1(y1)D2(z2)+y1y2z1D1(h)D2(z2), |
hz1[y1,y2,z2]=hy2z1D1(z2)D2(y1)−hz1z2D1(y2)D2(y1)−hy1z1D1(z2)D2(y2)+hz1z2D1(y1)D2(y2)+hy1z1D1(y2)D2(z2)−hy2z1D1(y1)D2(z2), |
−hz2[y1,y2,z1]=−hy2z2D1(z1)D2(y1)+hz1z2D1(y2)D2(y1)+hy1z2D1(z1)D2(y2)−hz1z2D1(y1)D2(y2)−hy1z2D1(y2)D2(z1)+hy2z2D1(y1)D2(z1). |
Thus, we get
y1[hy2,z1,z2]−y2[hy1,z1,z2]+hz1[y1,y2,z2]−hz2[y1,y2,z1]=0. |
We have studied transposed Poisson n-Lie algebras. We first established an important class of identities for transposed Poisson n-Lie algebras, which were subsequently used throughout the paper. We believe that the identities developed here will be useful in investigations of the structure of transposed Poisson n-Lie algebras in the future. Then, we introduced the notion of a strong transposed Poisson n-Lie algebra and derived an (n+1)-Lie algebra from a strong transposed Poisson n-Lie algebra with a derivation. Finally, we proved the conjecture of Bai et al. [2] for strong transposed Poisson n-Lie algebras.
The authors declare that they have not used artificial intelligence tools in the creation of this article.
Ming Ding was supported by the Guangdong Basic and Applied Basic Research Foundation (2023A1515011739) and the Basic Research Joint Funding Project of University and Guangzhou City under grant number 202201020103.
The authors declare that there is no conflict of interest.
[1] |
A. Abdessameud, A. Tayebi, On consensus algorithms for double-integrator dynamics without velocity measurements and with input constraints, Syst. Control Lett., 59 (2010), 812–821. https://doi.org/10.1016/j.sysconle.2010.06.019 doi: 10.1016/j.sysconle.2010.06.019
![]() |
[2] |
A. Abdessameud, A. Tayebi, On consensus algorithms design for double integrator dynamics, Automatica, 49 (2013), 253–260. https://doi.org/10.1016/j.automatica.2012.08.044 doi: 10.1016/j.automatica.2012.08.044
![]() |
[3] |
S. A. Ajwad, E. Moulay, M. Defoort, T. Menard, P. Coirault, Leader-following consensus of second-order multi-agent systems with switching topology and partial aperiodic sampled data, IEEE Control Syst. Lett., 5 (2021), 1567–1572. https://doi.org/10.1109/LCSYS.2020.3041566 doi: 10.1109/LCSYS.2020.3041566
![]() |
[4] |
M. Andreasson, D. V. Dimarogonas, H. Sandberg, K. H. Johansson, Distributed control of networked dynamical systems: Static feedback, integral action and consensus, IEEE T. Automat. Control, 59 (2014), 1750–1764. https://doi.org/10.1109/TAC.2014.2309281 doi: 10.1109/TAC.2014.2309281
![]() |
[5] |
R. Carli, A. Chiuso, L. Schenato, S. Zampieri, A PI consensus controller for networked clocks synchronization, IFAC Proc. Vol., 41 (2008), 10289–10294. https://doi.org/10.3182/20080706-5-KR-1001.01741 doi: 10.3182/20080706-5-KR-1001.01741
![]() |
[6] |
L. Cheng, Z. G. Hou, M. Tan, X. Wang, Necessary and sufficient conditions for consensus of double-integrator multi-agent systems with measurement noises, IEEE T. Automat. Control, 56 (2011), 1958–1963. https://doi.org/10.1109/TAC.2011.2139450 doi: 10.1109/TAC.2011.2139450
![]() |
[7] |
G. Dong, C. Yang, W. Zhu, Event-triggered average consensus of multiagent systems with switching topologies, Discrete Dyn. Nat. Soc., 2020 (2020), 8742014. https://doi.org/10.1155/2020/8742014 doi: 10.1155/2020/8742014
![]() |
[8] |
Y. Gao, L. Wang, Consensus of multiple double-integrator agents with intermittent measurement, Int. J. Robust Nonlin. Control, 20 (2010), 1140–1155. https://doi.org/10.1002/rnc.1496 doi: 10.1002/rnc.1496
![]() |
[9] |
Y. Gao, B. Liu, M. Zuo, T. Jiang, J. Yu, Consensus of continuous-time multiagent systems with general linear dynamics and nonuniform sampling, Math. Prob. Eng., 2013 (2013), 718759. https://doi.org/10.1155/2013/718759 doi: 10.1155/2013/718759
![]() |
[10] |
D. Goldin, Double integrator consensus systems with application to power systems, IFAC Proc. Vol., 46 (2013), 206–211. https://doi.org/10.3182/20130925-2-DE-4044.00023 doi: 10.3182/20130925-2-DE-4044.00023
![]() |
[11] |
T. A. Jesus, L. C. A. Pimenta, L. A. B. Tôrres, E. M. A. M. Mendes, Consensus for double-integrator dynamics with velocity constraints, Int. J. Control Autom. Syst., 12 (2014), 930–938. https://doi.org/10.1007/s12555-013-0309-0 doi: 10.1007/s12555-013-0309-0
![]() |
[12] | F. L. Lewis, H. Zhang, K. H. Movric, A. Das, Cooperative control of multi-agent systems, London: Springer, 2014. |
[13] |
Y. Li, C. Tan, A survey of the consensus for multi-agent systems, Syst. Sci. Control Eng., 7 (2019), 468–482. https://doi.org/10.1080/21642583.2019.1695689 doi: 10.1080/21642583.2019.1695689
![]() |
[14] | D. Liberzon, A. S. Morse, Basic problems in stability and design of switched systems, IEEE Control Syst., 19 (1999), 59–70. |
[15] |
P. Lin, Y. Jia, Further results on decentralised coordination in networks of agents with second-order dynamics, IET Control Theor. Appl., 3 (2009), 957–970. https://doi.org/10.1049/iet-cta.2008.0263 doi: 10.1049/iet-cta.2008.0263
![]() |
[16] |
H. Liu, G. Xie, L. Wang, Necessary and sufficient conditions for solving consensus problems of double-integrator dynamics via sampled control, Int. J. Robust Nonlin. Control, 20 (2010), 1706–1722. https://doi.org/10.1002/rnc.1543 doi: 10.1002/rnc.1543
![]() |
[17] | X. Liu, Y. Xie, F. Li, W. Gui, Consensus tracking of singular multiagent systems via sliding mode approach, IEEE T. Automat. Control, 2022, 1–7. https://doi.org/10.1109/TAC.2022.3214472 |
[18] |
X. Liu, Y. Xie, F. Li, W. Gui, Sliding-mode-based admissible consensus tracking of nonlinear singular multiagent systems under jointly connected opologies, IEEE T. Cybernetics, 52 (2022), 12491–12500. https://doi.org/10.1109/TCYB.2021.3081801 doi: 10.1109/TCYB.2021.3081801
![]() |
[19] |
C. Q. Ma, J. F. Zhang, Necessary and sufficient conditions for consensusability of linear multi-agent systems, IEEE T. Automat. Control, 55 (2010), 1263–1268. https://doi.org/10.1109/TAC.2010.2042764 doi: 10.1109/TAC.2010.2042764
![]() |
[20] | L. Qi, Some simple estimates for singular values of a matrix, Linear Algebra Appl., 56 (1984), 105–119. |
[21] |
J. Qin, H. Gao, A sufficient condition for convergence of sampled-data consensus for double-integrator dynamics with nonuniform and time-varying communication delays, IEEE T. Automat. Control, 57 (2012), 2417–2422. https://doi.org/10.1109/TAC.2012.2188425 doi: 10.1109/TAC.2012.2188425
![]() |
[22] |
J.Qin, Q. Ma, Y. Shi, L. Wang, Recent advances in consensus of multi-agent systems: a brief survey, IEEE T. Ind. Electron., 64 (2017), 4972–4983. https://doi.org/10.1109/TIE.2016.2636810 doi: 10.1109/TIE.2016.2636810
![]() |
[23] | W. J. Rugh, Linear system theory, New York: John Wiley and Sons, 1996. |
[24] |
A. Seuret, D. V. Dimarogonas, K. H. Johansson, Consensus of double integrator multi-agents under communication delay, IFAC Proc, Vol., 42 (2009), 376–381. https://doi.org/10.3182/20090901-3-RO-4009.00062 doi: 10.3182/20090901-3-RO-4009.00062
![]() |
[25] |
Z. G. Wu, Y. Xu, R. Lu, Y. Wu, T. Huang, Event-triggered control for consensus of multiagent systems with fixed/switching topologies, IEEE T. Syst. Man Cybernet. Syst., 48 (2018), 1736–1746. https://doi.org/10.1109/TSMC.2017.2744671 doi: 10.1109/TSMC.2017.2744671
![]() |
[26] | J. Zhan, X. Li, Consensus of multiple double-integrators with aperiodic sampled-data and switching topology, In: 34th Chinese Control Conference (CCC), 2015, 7113–7117. https://doi.org/10.1109/ChiCC.2015.7260765 |
[27] |
X. Y. Zhang, J. Zhang, Sampled-data consensus of multi-agent systems with general linear dynamics based on a continuous-time model, Acta Automat. Sin., 40 (2014), 2549–2555. https://doi.org/10.1016/S1874-1029(14)60400-6 doi: 10.1016/S1874-1029(14)60400-6
![]() |
[28] |
X. Y. Zhang, J. Zhang, Sampled-data consensus of general linear multi-agent systems under switching topologies: averaging method, Int. J. Control, 90 (2017), 275–288. https://doi.org/10.1080/00207179.2016.1177776 doi: 10.1080/00207179.2016.1177776
![]() |
[29] |
L. Zou, Z. D. Wang, D. H. Zhou, Event-based control and filtering of networked systems: a survey, Int. J. Autom. Comput., 14 (2017), 239–253. https://doi.org/10.1007/s11633-017-1077-8 doi: 10.1007/s11633-017-1077-8
![]() |
1. | K. Abdurasulov, F. Deraman, A. Saydaliyev, S. H. Sapar, Transposed Poisson Structures on Low Dimensional Quasi-Filiform Lie Algebras of Maximum Length, 2024, 45, 1995-0802, 5735, 10.1134/S1995080224606866 |