arxiv: 2605.00004 · v1 · submitted 2026-02-16 · 🧮 math.OC

Recognition: no theorem link

Optimization-based One-side Boundary Control of LWR Traffic Models

Eryn Vaid , Maria Teresa Chiri , Roberto Guglielmi , Gennaro Notomista

Authors on Pith no claims yet

Pith reviewed 2026-05-15 21:26 UTC · model grok-4.3

classification 🧮 math.OC

keywords LWR traffic modelboundary controlLyapunov functionalcontrol barrier functionconvex optimizationstabilityinvarianceconservation laws

0 comments

The pith

Sufficient conditions allow a convex optimization problem to select a single boundary control that both stabilizes traffic density and keeps it inside a safe set for generic LWR models.

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

The paper shows that, for any traffic flux function, there exist conditions under which the set of boundary controls that stabilize the system via a Lyapunov functional intersects the set of controls that enforce forward invariance via a barrier functional. A convex optimization problem then picks one feasible control from that intersection, achieving both goals with a single input applied at the boundary. This construction applies directly to one-dimensional conservation laws describing macroscopic traffic flow. If the conditions hold, the closed-loop density converges to the desired equilibrium while remaining inside the prescribed safe region. Simulations on standard flux functions illustrate that the resulting controller behaves as predicted.

Core claim

We determine sufficient conditions to ensure the existence of an optimal boundary control problem achieving both stability and invariance for a generic traffic flux function.

What carries the argument

The non-empty intersection between the Lyapunov-stabilizing controller set and the barrier-invariant controller set, from which a convex program selects the single boundary input.

If this is right

The closed-loop system is asymptotically stable to the target density equilibrium.
The prescribed safe subset of density states remains forward invariant under the applied control.
The same optimization framework works for any differentiable traffic flux function satisfying standard concavity assumptions.
Numerical solutions of the optimization at each time step produce a feedback law that simultaneously meets both objectives.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same Lyapunov-plus-barrier intersection idea could be tested on network-level traffic models by defining compatible boundary conditions at junctions.
If the optimization can be solved fast enough, the method supplies a real-time safety filter that can be layered on top of existing ramp-metering policies.
The approach provides an explicit certificate (the feasible set) that both stability and invariance are achieved, which could be used for formal verification of traffic control software.

Load-bearing premise

The intersection of the Lyapunov-stabilizing controller set and the barrier-invariant controller set is non-empty for the chosen traffic flux function, making the convex optimization feasible.

What would settle it

A concrete traffic flux function for which every boundary control that satisfies the Lyapunov decrease condition violates the barrier invariance condition (or vice versa), rendering the optimization problem infeasible.

Figures

Figures reproduced from arXiv: 2605.00004 by Eryn Vaid, Gennaro Notomista, Maria Teresa Chiri, Roberto Guglielmi.

**Figure 3.** Figure 3: For Example 1, this figure depicts the change in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

**Figure 4.** Figure 4: For Example 1, this figure displays the boundary [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗

read the original abstract

In this paper, we study the feasibility of a class of optimization-based boundary control of one-dimensional macroscopic traffic flow models, where stability and invariance are achieved by a single boundary control. We define the sets of controllers to stabilize the system to a desired state via Lyapunov functionals, and to ensure forward invariance of a desired subset via boundary control barrier functionals. The control input is then selected from the intersection of those sets via a convex optimization problem. We determine sufficient conditions to ensure the existence of an optimal boundary control problem achieving both stability and invariance for a generic traffic flux function. Simulation results showcase the behavior of the proposed optimization-based controller applied to conservation laws with several traffic flow functions.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

The paper gives a convex opt way to pick one boundary input that both stabilizes an LWR model and keeps density inside a safe set, but the feasibility claim for truly generic fluxes looks shaky without extra assumptions on f.

read the letter

The core contribution is a method that builds a Lyapunov set for asymptotic stability to a target density and a barrier set for forward invariance of a safe density interval, then solves a convex program to select a single boundary control from their intersection. This is applied to the LWR conservation law with one-sided actuation. The formulation is clean and directly addresses the practical constraint that only one boundary is available for control in many traffic settings. Simulations for several common flux functions show the closed-loop behavior, which is useful evidence that the scheme can be implemented numerically. What is new is the specific combination of the two functionals inside the optimization for generic f, rather than a new derivation of the PDE control theory itself. The approach sits squarely in the line of work on Lyapunov and barrier methods for hyperbolic systems. The main soft spot is the sufficient conditions for feasibility. The abstract states that the intersection is non-empty for a generic traffic flux, but the Lie derivatives at the controlled boundary depend on the local characteristic speed f'(ρ). For fluxes that are not strictly concave or monotone on the relevant density interval, the admissible control intervals from the Lyapunov decrease and the barrier condition can be disjoint, making the program infeasible. The stress-test note is right on this point; the paper would need either an explicit structural hypothesis on f or a proof that the sets overlap under the stated assumptions. Without that, the generality claim does not fully hold. This is worth sending to peer review. Readers working on boundary control of conservation laws or on traffic flow management will find the optimization framing practical and worth checking against their own models. The math is standard but the application is direct, so a referee can evaluate the derivations and the simulation details in one pass.

Referee Report

1 major / 2 minor

Summary. The manuscript develops an optimization-based one-sided boundary control strategy for the LWR traffic flow PDE. Controller sets are defined separately for Lyapunov stability to a desired equilibrium and for forward invariance of a safe set via barrier functionals; the input is chosen from their intersection by solving a convex optimization problem at each time. Sufficient conditions are stated for this intersection to be nonempty (hence for feasibility) when the flux function is generic, with the approach illustrated by numerical simulations on several common traffic flux functions.

Significance. If the claimed sufficient conditions are valid, the work supplies a unified, computationally tractable method for simultaneously enforcing stability and safety constraints on a class of scalar conservation laws arising in traffic modeling. The combination of Lyapunov and barrier certificates inside a convex program is a natural extension of finite-dimensional CBF-CLF techniques to boundary-controlled hyperbolic PDEs and could be useful for real-time traffic management applications.

major comments (1)

[Abstract and main theorem on sufficient conditions] Abstract and the statement of sufficient conditions: the claim that the intersection of the Lyapunov-stabilizing controller set and the barrier-invariant controller set is nonempty for a generic C² flux function f is not supported without additional structural hypotheses. The Lie derivatives of V and B at the controlled boundary contain the term f(ρ(0,t)) or f(ρ(L,t)), whose admissible range for decrease/invariance depends on the sign and magnitude of f'(ρ) on the relevant density interval. For fluxes lacking strict concavity or monotonicity, these admissible intervals for u can be disjoint, rendering the convex program infeasible and contradicting the generic statement.

minor comments (2)

[Sufficient conditions section] The manuscript would benefit from an explicit statement (perhaps in a remark or corollary) of the minimal extra assumptions on f (e.g., f'' < 0 on the operating interval) that guarantee the intersection is nonempty.
[Numerical results] Simulation figures should include quantitative metrics (e.g., time to enter the invariant set, L²-norm decay rate of the Lyapunov function) rather than qualitative density plots alone.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the careful reading and insightful comments on our manuscript. The main concern is the support for the sufficient conditions ensuring a nonempty intersection of the controller sets for generic flux functions. We respond point-by-point below and will revise the manuscript accordingly.

read point-by-point responses

Referee: [Abstract and main theorem on sufficient conditions] Abstract and the statement of sufficient conditions: the claim that the intersection of the Lyapunov-stabilizing controller set and the barrier-invariant controller set is nonempty for a generic C² flux function f is not supported without additional structural hypotheses. The Lie derivatives of V and B at the controlled boundary contain the term f(ρ(0,t)) or f(ρ(L,t)), whose admissible range for decrease/invariance depends on the sign and magnitude of f'(ρ) on the relevant density interval. For fluxes lacking strict concavity or monotonicity, these admissible intervals for u can be disjoint, rendering the convex program infeasible and contradicting the generic statement.

Authors: We acknowledge the referee's observation that the admissible intervals for the boundary input u are determined by the sign and magnitude of f'(ρ) evaluated at the controlled boundary. Our sufficient conditions in the manuscript are derived under the standard structural assumptions for LWR traffic flux functions: f ∈ C², f(0) = f(ρ_max) = 0, f'(ρ) > 0 on [0, ρ_c) and f'(ρ) < 0 on (ρ_c, ρ_max], together with strict concavity (f'' < 0). These properties ensure that the sets defined by the Lyapunov decrease condition and the barrier invariance condition have nonempty intersection for all states in the relevant domain. The term 'generic' in the paper is used to indicate that the result holds for the broad class of fluxes satisfying these traffic-model hypotheses (which form a dense set in the appropriate function space). Nevertheless, to eliminate any ambiguity, we will revise the abstract and the statement of the main theorem to list these structural hypotheses explicitly rather than relying on the word 'generic' alone. This clarification will also include a brief remark on why the concavity and monotonicity properties prevent the admissible intervals from becoming disjoint. revision: yes

Circularity Check

0 steps flagged

No circularity in derivation chain

full rationale

The paper applies standard Lyapunov functionals and boundary control barrier functionals from control theory to the LWR PDE, then selects the control from the intersection of the resulting sets via a convex program. Sufficient conditions for feasibility are stated for a generic flux function without any parameter fitting that defines the result, without self-citation load-bearing the central claim, and without renaming or smuggling ansatzes. All steps remain independent of the target existence result itself.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claim rests on the existence of suitable Lyapunov and barrier functionals for the LWR system and on the non-emptiness of their controller-set intersection; these are standard domain assumptions in boundary control of hyperbolic conservation laws rather than new entities or fitted parameters.

axioms (2)

domain assumption Existence of Lyapunov functionals that certify stability to a desired equilibrium for the LWR PDE under boundary control
Invoked to define the stabilizing controller set
domain assumption Existence of boundary control barrier functionals that certify forward invariance of a desired density subset
Invoked to define the invariance controller set

pith-pipeline@v0.9.0 · 5414 in / 1278 out tokens · 21210 ms · 2026-05-15T21:26:26.960942+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

23 extracted references · 23 canonical work pages

[1]

IEEE Control Systems Letters , year =

Maria Teresa Chiri and Roberto Guglielmi and Gennaro Notomista , title =. IEEE Control Systems Letters , year =

work page
[2]

Ardekani and Mostafa Ghandehari and Shiva M

Siamak A. Ardekani and Mostafa Ghandehari and Shiva M. Nepal , title =. BULETINUL INSTITUTULUI POLITEHNIC DIN IAŞI , year =

work page
[3]

M. J. Lighthill and G. B. Whitham , title =. Proceedings of the Royal Society of London. Series A, Mathematical and Physical Sciences , year =

work page
[4]

P. I. Richards , title =. Operations Research , year =

work page
[5]

2007 , publisher=

Control and nonlinearity , author=. 2007 , publisher=

work page 2007
[6]

IEEE Transactions on Automatic control , volume=

A strict Lyapunov function for boundary control of hyperbolic systems of conservation laws , author=. IEEE Transactions on Automatic control , volume=. 2007 , publisher=

work page 2007
[7]

Archive for rational mechanics and analysis , volume=

Rigorous derivation of nonlinear scalar conservation laws from follow-the-leader type models via many particle limit , author=. Archive for rational mechanics and analysis , volume=. 2015 , publisher=

work page 2015
[8]

Regularity and

Blandin, Sébastien and Litrico, Xavier and Delle Monache, Maria Laura and Piccoli, Benedetto and Bayen, Alexandre , journal=. Regularity and. 2017 , volume=

work page 2017
[9]

2022 , publisher=

Control problems for conservation laws with traffic applications: modeling, analysis, and numerical methods , author=. 2022 , publisher=

work page 2022
[10]

IEEE control systems letters , volume=

Integral control barrier functions for dynamically defined control laws , author=. IEEE control systems letters , volume=. 2020 , publisher=

work page 2020
[11]

Proceedings of 35th IEEE conference on decision and control , volume=

Control Lyapunov functions: New ideas from an old source , author=. Proceedings of 35th IEEE conference on decision and control , volume=. 1996 , organization=

work page 1996
[12]

Communications in partial differential equations , volume=

First order quasilinear equations with boundary conditions , author=. Communications in partial differential equations , volume=. 1979 , publisher=

work page 1979
[13]

Hu, Hanjiang and Liu, Changliu , journal=. On the

work page
[14]

On kinematic waves

Lighthill, Michael James and Whitham, Gerald Beresford , journal=. On kinematic waves. 1955 , publisher=

work page 1955
[15]

Operations research , volume=

Shock waves on the highway , author=. Operations research , volume=. 1956 , publisher=

work page 1956
[16]

2010 , publisher=

Optimal control of partial differential equations: theory, methods, and applications , author=. 2010 , publisher=

work page 2010
[17]

Optimal control of the

Daudin, Samuel , journal=. Optimal control of the. 2023 , publisher=

work page 2023
[18]

Ji Wang and Miroslav Krstic , year=. Output-. 2309.05596 , archivePrefix=

work page arXiv
[19]

Li, Zongyi and Kovachki, Nikola Borislavov and Azizzadenesheli, Kamyar and Bhattacharya, Kaushik and Stuart, Andrew and Anandkumar, Anima and others , booktitle=. Fourier

work page
[20]

2019 18th European control conference (ECC) , pages=

Control barrier functions: Theory and applications , author=. 2019 18th European control conference (ECC) , pages=. 2019 , organization=

work page 2019
[21]

Ledyaev, Yuri S and Sontag, Eduardo D , journal=. A. 1999 , publisher=

work page 1999
[22]

2410.15146 , archivePrefix=

Rafael Vazquez and Jean Auriol and Federico Bribiesca-Argomedo and Miroslav Krstic , year=. 2410.15146 , archivePrefix=

work page arXiv
[23]

Comparison of the one-equation LWR models for density and for speed , volume =

Maciejewski, Marek , year =. Comparison of the one-equation LWR models for density and for speed , volume =. IOP Conference Series: Materials Science and Engineering , doi =

work page