Morphisms of tautological control systems

Qianqian Xia

REVIEW 4 major objections 4 minor 14 references

Morphisms of tautological control systems

T0 review · 4 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper establishes sufficient conditions under which the reachability of a tautological control system on a base manifold is inherited by a lifted system on another manifold, via trajectory-preserving morphisms along proper…

desk verdict A plausible extension of Grasse-style lifting to tautological control systems, with genuinely new real-analytic and second-order results, but the main reachability theorem depends on an omitted proof and an unproven transfer of global-in-time from ordinary systems. read the letter →

arxiv 1908.03562 v1 pith:ZCLUHUUM submitted 2019-08-08 math.OC

Qianqian Xia This is my paper · ORCID

classification math.OC MSC 93B0393B1793C10

keywords TautologicalcontrolsystemMorphismReachabilityTrajectoryliftingEtaleopen-loopSecond-ordertypeSheaftheoryPropersubmersion

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Tautological control systems describe dynamics as sheaves of vector fields rather than as equations with a chosen control parameterization, so local and global questions can be separated cleanly. This paper asks when such a system on a base manifold can be pulled back to another manifold through a smooth map $\Phi: M \to N$, preserving trajectories, and when reachability of the base system forces reachability of the lifted one. The main theorem says that if the base system is globally generated by pointwise linearly independent vector fields, $\Phi$ is a proper submersion (smooth map with surjective derivative and compact fibers), the base is reachable from a point $y_0$ by piecewise constant etale open-loop trajectories, and the fiber $\Phi^{-1}(y_0)$ is connected, then a trajectory-preserving lifted system exists and is reachable from every point of that fiber. A sympathetic reader should care because this gives a global, control-parameter-free analogue of classical trajectory lifting, with smooth and real-analytic versions, and it improves the input count for second-order mechanical systems.

What carries the argument

The central object is the trajectory-preserving morphism $(\Phi, \Phi^\#)$: a map $\Phi \in C^r(M,N)$ together with continuous linear maps $\Phi^\#_V$ that send every local section $Y$ of the base presheaf over an open set $V \subseteq N$ to a vector field on $\Phi^{-1}(V)$ satisfying $T_x\Phi(\Phi^\#(Y)(x)) = Y(\Phi(x))$ at each $x$. This tangent identity is exactly what makes integral curves of the lifted field project to integral curves of $Y$. The proof's second ingredient is the presheaf of vertical fields $\mathcal{H}(U)=\{X \in \Gamma^\nu(TU) \mid X(x) \in \ker T_x\Phi\}$; adding $\mathcal{H}$ to the lifted presheaf lets trajectories run freely inside the fibers, so a connected fiber becomes a reachability set. The etale open-loop subsystem, built from locally integrable sections of the sheaf of time-varying vector fields, is the trajectory class in which reachability is measured.

What would settle it

Build an explicit example of a proper submersion $\Phi: M \to N$ and a tautological control system $H$ on $N$, globally generated by independent vector fields, for which the trajectory-preserving lift $\Phi^\#(Y)$ of some $Y$ has an integral curve that dies before its projection through $\Phi(x)$. If such an example exists, Theorem 3 fails, because its proof needs the morphism to be global in time; if none exists, proving the equal-maximal-interval property would complete the argument.

Watch

Extended reading notes

Core claim

The core discovery is that reachability passes upward along a trajectory-preserving morphism when three conditions hold: the base system's reachable set covers the image of the map, the morphism is global in time, and each fiber is a reachability set. Theorem 3 realizes these conditions for proper submersions by adjoining vertical vector fields tangent to the fibers: connectedness of $\Phi^{-1}(y_0)$ makes that fiber path connected, hence a reachability set for the vertical subsystem, while properness of $\Phi$ is used to ensure the lifted morphism is global in time via a corollary imported from the classical lifting theory. With the fiber a reachability set and the base reachable from $y_0$, Proposition 5 yields reachability of the lifted system from every $x_0 \in \Phi^{-1}(y_0)$ by its piecewise constant etale open-loop subsystem. The same construction also yields small-time local controllability, and for second-order type systems the lifted system can be chosen within the same class with fewer inputs than the local classical construction.

Load-bearing premise

The entire reachability-lifting argument assumes without proof that a trajectory lifted through a proper submersion $\Phi: M \to N$ exists exactly as long as its projection, a 'global in time' fact imported from the classical setting of $C^1$ control systems and never proved for the sheaf-theoretic etale trajectories used here.

Editorial extensions

If this is right

Under the hypotheses of Theorem 3, from any single lift $x_0$ of a reachable base point $y_0$, the lifted system reaches every point of $M$; reachability is inherited fiber-by-fiber, not just pointwise.
When the vertical distribution is globally finitely generated (the smooth case), the lifted system can be chosen globally generated, so the result applies to ordinary control systems through the correspondence established in Propositions 7 and 8.
Small-time local controllability of the base system from $y_0$ lifts to small-time local controllability of the lifted system from every point of the connected fiber (Proposition 6).
For second-order type systems, the lifted system can be chosen second-order type with $n-m+k$ inputs (fiber dimension plus base controls) rather than $2n-2m+k$, and the construction gives a globally defined system rather than only a local one (Theorem 5 and Remark 7).

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

One could replace the imported properness condition by an explicit 'equal maximal interval' hypothesis on the lifted morphism; the reachability argument would then go through unchanged and would no longer depend on the unproved transfer from $C^1$ systems.
The same construction suggests a recipe for observer or synchronization designs: whenever a projection has connected fibers and the base system is reachable, augmenting the lifted dynamics with arbitrary vertical fields should make the whole fiber reachable regardless of the control parametrization.
For real-analytic systems, the failure of global finite generation for the vertical distribution suggests that global reachability lifting may be controlled by a sheaf-cohomology obstruction; checking whether vanishing of the relevant cohomology restores the global result would be a natural next step.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

The paper extends Grasse's lifting results to Lewis's tautological control systems, and that is a real contribution. The sheaf framework is a natural fit for the local-global distinction that Grasse's local construction leaves unresolved, and the real-analytic version via connections (Theorem 2) is genuinely new. The second-order type application is also a real plus: the control count drops from 2n−2m+k to n−m+k, and the lifted system is global rather than local. If the main theorem holds, it is a useful addition to the tautological control systems literature, though not a field-shifting one.

The main theorem is plausible, but the proof as written has three soft spots. First, Proposition 5 is the load-bearing reachability transfer, and its proof is omitted with a reference to an analogous result in [7]. In a paper whose central theorem depends on that proposition, an omitted proof is a real gap. Second, Theorem 3 imports 'global in time' from Corollary 1 of [7], which is proved for ordinary C^1 control systems. No argument shows that the maximal-interval equality J_{Φ#(Y)}(s,x)=J_Y(s,Φ(x)) carries over to tautological systems with etale sheaf trajectories under properness. This is exactly the step where the reachability transfer could fail, so it cannot be taken on faith. Third, the construction of Φ# in Theorem 1 uses a partition of unity chosen anew for each Y. As written, that does not obviously give the linear map required by Definition 3, nor even a well-defined map on restricted sections. This particular issue is fixable by fixing a splitting of TΦ and a fixed partition of unity, but the paper does not do that.

None of these are demonstrated contradictions; they are missing arguments at critical points. The framework and the statements of the theorems are coherent, and the author clearly knows the literature. Proposition 7 is essentially a definitional equivalence, but it is harmless. The examples, especially the Möbius band, are illustrative.

Who gets value from this? Researchers working on tautological control systems, lifting theory, or real-analytic geometric control. A serious referee should engage with it, but the verdict should be 'revise' with a demand for a full proof of Proposition 5, a proof of the global-in-time transfer, and a repaired construction of Φ#. If those are supplied, the paper would be a solid, if niche, addition. As it stands, the main theorem is not fully proved.

Referee Report

4 major / 4 minor

Summary. The paper studies morphisms of tautological control systems in the framework introduced by Lewis. Given a C^ν tautological control system H on a manifold N and a map Φ : M → N, the author asks (i) when there exists a tautological control system G on M with a trajectory-preserving morphism (Φ, Φ#) from G to H, and (ii) when reachability of H implies reachability of G. The main result, Theorem 3, asserts that if H is globally generated by linearly independent vector fields, Φ is a proper submersion, y0 ∈ Φ(M), Φ(M) is reachable from y0 by piecewise constant etale open-loop trajectories, and the fiber Φ^{-1}(y0) is connected, then there exists a tautological control system G on M, with a trajectory-preserving morphism, that is reachable from every point of the fiber. Section 3 also contains a real-analytic existence theorem (Theorem 2), a small-time local controllability result (Proposition 6), and a discussion of the relation between lifting and morphisms (Propositions 7–8). Section 4 applies these ideas to second-order type control systems.

Significance. If the main results are correct, the paper would extend Grasse's lifting theorem for ordinary C^1 control systems to tautological control systems with sheaf-theoretic and etale trajectories, while also addressing global constructions and some real-analytic cases. The framework is potentially valuable because it distinguishes local and global issues and avoids a fixed parametrisation by controls. The paper is clearly organized and honestly states limitations in Remarks 3 and 5. However, the proof of the central reachability theorem depends on several unproved or insufficiently justified assertions, including an omitted proof of Proposition 5, an unjustified transfer of a global-in-time property from ordinary control systems, and a construction of the morphism in Theorem 1 that may not satisfy the linearity requirement in Definition 3. These gaps must be repaired before the main claims can be accepted.

major comments (4)

[Section 3.2, Proposition 5] The proof of Proposition 5 is omitted entirely, with the sentence 'The proof is analogous to the proof for Theorem 5 in [7] so we omit it here.' This proposition is the key step that turns a trajectory-preserving morphism into a reachability implication for the lifted system, and Theorem 3 invokes it directly. The tautological/etale setting differs from the ordinary C^1 setting of Grasse's Theorem 5, so an analogy is not a proof. A complete proof must be supplied, showing in particular how the reachability of H by the piecewise constant etale open-loop subsystem transfers to reachability of G under the stated assumptions.
[Section 3.2, proof of Theorem 3] The proof states: 'For Φ is a proper mapping, by Corollary 1 in [7], we know the trajectory-preserving morphism from G to H is global in time.' Corollary 1 of Grasse is proved for ordinary C^1 control systems, whereas Definition 5 and Proposition 5 are formulated for C^ν tautological systems with etale sheaf trajectories. No argument is given that the maximal-interval equality J_{Φ#(Y)}(s,x)=J_Y(s,Φ(x)) holds for etale trajectories under properness of Φ, and no bridge is provided through Propositions 3–4 and 7–8. Without global-in-time, Proposition 5 cannot be applied, so the reachability conclusion of Theorem 3 is not established.
[Section 3.1, Theorem 1] The construction of Φ# defines, for each Y ∈ G(N), a vector field X by patching local lifts with a partition of unity. This does not demonstrate the linearity of the resulting map Y ↦ Φ#(Y) required in Definition 3(ii)(a), which calls for a family L_V of continuous linear mappings. Moreover, the proof only constructs Φ# for global sections in G(N), not for sections over arbitrary open V ⊆ N, and no extension to a compatible family L_V is provided. Since Theorem 3 depends on Theorem 1 for the existence of a trajectory-preserving morphism, this gap is load-bearing.
[Section 3.2, Proposition 6] In the proof of Proposition 6, the sentence 'By the mapping Φ#, we know that x is reachable from some point \bar{x} ∈ Φ^{-1}(y0) in time at most t' lifts a trajectory of H to a trajectory of G without invoking global-in-time or Proposition 5. This step is not justified; reachability of H does not automatically imply reachability of G unless the trajectory-preserving morphism is global in time and the fiber reachability assumption holds. Proposition 6 is a stated result, so its proof must be completed or its hypotheses strengthened.

minor comments (4)

[Abstract and general] The manuscript contains frequent typographical errors and broken hyphenation (e.g., 'sy s-tems' in the abstract, 'lea ds' in the abstract). A careful proofreading pass is needed.
[Theorem 2] The statement 'Let Φ : M → N be a real analytic vector bundle' is unclear: a vector bundle is typically a triple (E, π, M), not a mapping between two manifolds. Please clarify whether Φ is a vector bundle projection or a submersion with additional vector-bundle structure.
[Section 3.3, Proposition 7] In the proof of Proposition 7, the well-definedness of Φ#_V on the subset {F^u_2|V : u ∈ C_2} does not by itself give a continuous linear map on Γ^ν(TV). The argument needs to construct an extension to the whole space and verify the linearity and compatibility conditions of Definition 3(ii).
[Section 4, Theorem 5] The hypothesis 'T(φ(P)) ⊆ R_{HΣ2}((z0,0), G_{HΣ2}, pwc)' should be stated more precisely; as written it is ambiguous whether the reachable set is taken in the tangent bundle TQ or in the image φ(P).

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; reliance on Grasse's global-in-time result is an external transfer concern, not a circular reduction.

full rationale

The paper's main derivation is not circular. Theorem 1 constructs a lifted tautological system G by partition of unity and directly verifies the trajectory-preserving condition T_xΦ(Φ#Y)=Y(Φ(x)) from Proposition 1, so the existence of the morphism is an actual construction rather than an assumption. Theorem 3 then augments G with vertical fields tangent to ker TΦ, proves directly that a connected fiber Φ^{-1}(y0) is a reachability set for G using the definition of trajectories, and combines this with the assumed reachability of H to conclude reachability of G. The global-in-time property of the morphism is imported from Corollary 1 of Grasse [7] rather than proved in the tautological/etale setting, and Proposition 5's proof is omitted as 'analogous to Theorem 5 in [7]'. These are external-support and proof-transfer concerns, not circularity: [7] is not authored by the present author, and the cited result is not identical to the conclusion being derived. Proposition 7 is a definitional correspondence between liftability of control systems and trajectory-preserving morphisms, but it does not feed into the main reachability theorem and is not used to derive a conclusion from itself. There are no fitted inputs called predictions, no uniqueness theorem imported from the authors' own prior work, and no known result merely renamed as a derivation. Accordingly, no circular step can be exhibited.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The paper introduces no free parameters and no new postulated entities. It relies on several external results: partitions of unity, finite generation of smooth distributions, the global-in-time property imported from Grasse's ordinary control system theory, and Cartan's Theorem A for the analytic discussion. The most fragile dependence is the transferred global-in-time property.

assumptions (5)

standard math Partitions of unity exist on second-countable Hausdorff smooth manifolds.
Used in Theorem 1 and Proposition 9 to patch local lifts into global vector fields.
standard math Connected smooth manifolds are path connected.
Used in Theorem 3 and Theorem 5 to connect points in fibers by piecewise smooth paths.
domain assumption Smooth generalized distributions are globally finitely generated (Theorem 4 of [12]).
Used to replace the sheaf H by a globally generated presheaf ~H in the construction following Theorem 3.
domain assumption A proper submersion yields a global-in-time trajectory-preserving morphism (imported from Corollary 1 of [7]).
Assumed in Theorem 3 and Proposition 6 to apply Proposition 5; no proof is given in the tautological framework.
domain assumption Real analytic vector bundles admit real analytic linear connections.
Used in Theorem 2 to define horizontal lifts of vector fields.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Morphisms of tautological control systems." pith.science (2026). https://pith.science/paper/ZCLUHUUM

@misc{pith2026190803562,
  author       = {Pith},
  title        = {Pith review of: Morphisms of tautological control systems},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ZCLUHUUM}},
  note         = {Machine review of arXiv:1908.03562}
}

abstract

In this paper, we investigate morphisms of tautological control systems. Given a tautological control system $\mathfrak{H}$ on the manifold N and a mapping $\Phi: M \to N$, we study existence of tautological control system $\mathfrak{G}$ on the manifold $M$ such that there exists a trajectory-preserving morphism $(\Phi, \Phi^ #)$ from $\mathfrak{G}$ to $\mathfrak{H}$. Sufficient conditions are given such that reachability of $\mathfrak{H}$ implies the reachability of $\mathfrak{G}$. Correspondence between the notion of lifting ordinary control systems and morphisms of tautological control systems are examined. We give an application of the above results to the class of second-order type control systems, where the special structure of second-order type leads to additional results.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 14 canonical work pages

[7]

Grasse KA (2015) Controllability properties of nonlinea r systems induced by lifting. Math. Control Signals Syst 27(2):187-217

work page 2015
[1]

Krener AJ (1977) A decomposition theory for diﬀerentiabl e systems. SIAM J. Control Optim 15(5):813-829

work page 1977
[2]

IEEE Trans

Grizzle JW, Marcus SI (1985) The structure of nonlinear co ntrol systems possessing symmetries. IEEE Trans. Automat. Control 30(3):248-258

work page 1985
[3]

IEEE Trans

Pappas GJ, Laﬀerriere G, Sastry S (2000) Hierarchically c onsistent control systems. IEEE Trans. Automat. Control 45(6):1144-1160

work page 2000
[4]

IEEE Trans Autom Control 47(5):745-756

Pappas GJ, Simic S (2002) Consistent abstractions of aﬃne control systems. IEEE Trans Autom Control 47(5):745-756

work page 2002
[5]

Tabuada P, Pappas GJ (2005) Quotients of fully nonlinear c ontrol systems. SIAM J. Control Optim 43(5):1844-1866

work page 2005
[6]

Reports on Math- ematical Physics 72(1):103-120

Xia Q, Geng Z (2013) Quotients of aﬃne connection control s ystems. Reports on Math- ematical Physics 72(1):103-120

work page 2013
[8]

Springer B riefs in Electrical and Computer Engineering-Control, Automation and Robotics

Lewis AD (2014) Tautological Control Systems. Springer B riefs in Electrical and Computer Engineering-Control, Automation and Robotics. S pringer-Verlag, New York- Heidelberg-Berlin

work page 2014

Show all 14 references

[9]

Sussmann HJ (1973) Orbits of families of vector ﬁelds and i ntegrability of distributions. Trans. Amer. Math. Soc 180(1):171

1973
[10]

2nd edition

Bredon GE (1997) Sheaf Theory. 2nd edition. Graduate Tex ts in Mathematics 170. Springer-Verlag, New York-Heidelberg-Berlin

1997
[11]

Springer Briefs in Mathematics

Jafarpour S, Lewis AD (2014) Time-Varying Vector Fields and Their Flows. Springer Briefs in Mathematics. Springer-Verlag, New York-Heidelb erg-Berlin

2014
[12]

Smooth dist ributions are ﬁnitely generated

Drager LD, Lee JM, Park E (2012) Richardson K. Smooth dist ributions are ﬁnitely generated. Annals of Global Analysis and Geometry 41(3):35 7-369

2012
[13]

Colloque tenu a Bruxelles:41-55

Cartan H (1953) Varietes analytiques complexes et cohom ologie. Colloque tenu a Bruxelles:41-55

1953
[14]

Modeling, Analysis and Design for Simple Mechanical Control Systems

Bullo F, Lewis AD (2004) Geometric Control of Mechanical Systems. Modeling, Analysis and Design for Simple Mechanical Control Systems. Springer , New York

2004

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.

[1] [7]

Grasse KA (2015) Controllability properties of nonlinea r systems induced by lifting. Math. Control Signals Syst 27(2):187-217

work page 2015

[2] [1]

Krener AJ (1977) A decomposition theory for diﬀerentiabl e systems. SIAM J. Control Optim 15(5):813-829

work page 1977

[3] [2]

IEEE Trans

Grizzle JW, Marcus SI (1985) The structure of nonlinear co ntrol systems possessing symmetries. IEEE Trans. Automat. Control 30(3):248-258

work page 1985

[4] [3]

IEEE Trans

Pappas GJ, Laﬀerriere G, Sastry S (2000) Hierarchically c onsistent control systems. IEEE Trans. Automat. Control 45(6):1144-1160

work page 2000

[5] [4]

IEEE Trans Autom Control 47(5):745-756

Pappas GJ, Simic S (2002) Consistent abstractions of aﬃne control systems. IEEE Trans Autom Control 47(5):745-756

work page 2002

[6] [5]

Tabuada P, Pappas GJ (2005) Quotients of fully nonlinear c ontrol systems. SIAM J. Control Optim 43(5):1844-1866

work page 2005

[7] [6]

Reports on Math- ematical Physics 72(1):103-120

Xia Q, Geng Z (2013) Quotients of aﬃne connection control s ystems. Reports on Math- ematical Physics 72(1):103-120

work page 2013

[8] [8]

Springer B riefs in Electrical and Computer Engineering-Control, Automation and Robotics

Lewis AD (2014) Tautological Control Systems. Springer B riefs in Electrical and Computer Engineering-Control, Automation and Robotics. S pringer-Verlag, New York- Heidelberg-Berlin

work page 2014