Searcharxiv⌕ Search

arXiv · 2610.24080

An equiregular counterexample to the Sard conjecture for sub-Riemannian endpoint maps

Abstract

Lerario, Rizzi and Tiberio disproved the sub-Riemannian Sard conjecture with a rank-two distribution on $\mathbb R^4$, of degree of nonholonomy five at the origin, whose endpoint map from that point has no regular values. Their distribution is neither equiregular nor real-analytic, and the equiregular case has remained open. We construct an equiregular rank-two distribution on an open set diffeomorphic to $\mathbb R^9$, with small growth vector $(2,3,5,8,9)$, whose endpoint map from the origin has no regular values. This growth vector is maximal and generic for rank-two distributions in dimension nine, so the failure of the Sard property is not tied to any degeneracy of the growth vector. Singular curves of the Lerario--Rizzi--Tiberio-type structure built from the same coefficient lift, inside the strip where our distribution is defined, to singular curves, and the vertical displacement of a horizontal loop is an explicit strip integral. Realising an open set of vertical displacements by singular loops reduces to a graded moment problem, which we solve with pairs of bumps of opposite winding, an anisotropic rescaling, and an inverse function theorem uniform over infinitely many scales. We also state, with proofs deferred to a planned longer version, two results about our distribution: a description of its abnormal dynamics in the cotangent bundle, which explains the example from a dynamical point of view, and the non-constancy of its Tanaka symbol. The latter motivates an equiregular counterexample in dimension twelve with constant Tanaka symbol (constant nilpotentisation), announced here and to be proved in a separate paper. The longer version will also contain obstructions to real-analytic and lower-dimensional versions, and a transversal variant of the construction, which also leads to a modification of the Lerario--Rizzi--Tiberio counterexample.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Bangjie Wang, Igor Zelenko. 2026-10-16. An equiregular counterexample to the Sard conjecture for sub-Riemannian endpoint maps. https://arxiv.org/abs/2610.24080

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

Metric Lines in Engel-type Groups

In the framework of sub-Riemannian manifolds, a natural question is: What are the ''metric lines'' (i.e., isometric embeddings of the real line)? This article presents a conjecture classifying metric lines in Carnot groups and investigates the question for Carnot groups of arbitrary rank. It focuses on metric lines in Engel-type groups, whose sub-Riemannian structure is defined on a distribution of rank $n+1$. The approach employs a novel technique, the sequence method, initially developed to study metric lines in jet space.

math.DG↗

Ancient Gauss curvature flows of bounded width

In this paper, we construct a pancake-like ancient compact solution with flat sides to the Gauss curvature flow, contained in a slab. Also, we construct sausage-like ancient compact solutions to the $α$-Gauss curvature flow with $α>\frac{1}{2}$, asymptotic to a round cylinder.

math.DG↗

Metric Lines in the Space of Curves

This paper investigates sub-Riemannian geodesics in the jet space of curves. We establish the existence of two distinct families of metric lines, that is, globally minimizing geodesics, in the $2$-jet space of plane curves. This result contributes to the broader classification of metric lines in jet spaces. Additionally, we present precise criteria that characterize when a sub-Riemannian geodesic in the $2$-jet space of plane curves is a metric line.

math.DG↗