Additive kinematic formulas for flag area measures (Q832496)

From MaRDI portal
scientific article
Language Label Description Also known as
English
Additive kinematic formulas for flag area measures
scientific article

    Statements

    Additive kinematic formulas for flag area measures (English)
    0 references
    0 references
    0 references
    25 March 2022
    0 references
    Let \(G\) be a subgroup of \(\operatorname{O}(n)\) and let \(H\) be a closed subgroup of \(G\cap \operatorname{O}(n-1)\), where \(\operatorname{O}(n-1)\subset \operatorname{O}(n)\) is the stabilizer of the first basis vector. In the following, \(\operatorname{Area}_{G/H}^G\) denotes the space of smooth flag area measures that are equivariant with respect to \(G\). These objects are translation invariant valuations that take values in the space of signed measures on \(G/H\) and that can be represented by certain smooth differential forms. In their main result, the authors show that \(\operatorname{Area}_{G/H}^G\) is finite-dimensional if and only if \(G\) acts transitively on the unit sphere and that in this case there exist local additive kinematic formulas of the form \[\int_G \Phi_i(K+g L,\kappa\cap g \lambda) d g = \sum_{k,l=1}^N c_{k,l}^i \Phi_k(K,\kappa) \Phi_l(L,\lambda).\] Here, \(\{\Phi_1,\ldots,\Phi_N\}\) is a basis of \(\operatorname{Area}_{G/H}^G\), \(\kappa,\lambda\) are Borel subsets of \(G/H\) and \(K\) and \(L\) are convex bodies. Note that these formulas can be encoded by a cocommutative, coassociative coproduct. For \(G\subset \operatorname{SO}(n)\) the elements of \(\operatorname{Area}_{G/\{1\}}^G\) are also called rotation measures and an explicit construction and classification of such measures is given. Furthermore, the authors give an algebraic description of the kinematic formulas for \(\operatorname{Area}_{G/H}^G\) when \(H\subset \operatorname{SO}(n-1)\) is a closed subgroup. In particular, an explicit description of this algebraic structure is given for the important case \(G=\operatorname{SO}(n)\), \(H=S(\operatorname{O}(p)\times \operatorname{O}(q))\), \(p+q=n-1\), in which \(G/H=\operatorname{Flag}_{1,p+1}\) is the incomplete flag manifold consisting of pairs \((v,E)\), where \(E\) is a \((p+1)\)-plane and \(v\in E\). Note that the results of this paper generalize previous results by [\textit{T. Wannerer}, Adv. Math 263, 1--44 (2014; Zbl 1296.53150)] and [\textit{A. Bernig}, Geom. Dedicata 203, 85--110 (2019; Zbl 1432.53099)].
    0 references
    flag area measures
    0 references
    kinematic formulas
    0 references
    valuations
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references