🔢 mathematics

Hybrid Systems as Coalgebras: Lyapunov Morphisms for Zeno Stability

This paper unifies various stability concepts for hybrid dynamical systems, including the novel notion of Zeno stability, by modeling them as coalgebras and demonstrating that Lyapunov functions are morphisms into specific target systems that define the desired stability behavior.

Original authors: Joe Moeller, Aaron D. Ames

Published 2026-04-08

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Joe Moeller, Aaron D. Ames

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to understand how a complex machine works. Maybe it's a robot that walks, a car that brakes, or a ball that bounces. These machines don't just move smoothly like a car on a highway; they also have sudden "jumps." A walking robot swings its leg (smooth), then slams it into the ground (sudden jump). A bouncing ball falls (smooth), then instantly reverses direction when it hits the floor (sudden jump).

In math and engineering, we call these Hybrid Systems. They are a mix of "flow" (smooth movement) and "jumps" (instant changes).

The big problem scientists face is: How do we know if these machines are stable?

Will the robot fall over?
Will the bouncing ball eventually stop, or will it bounce forever?
Will the ball bounce so many times in such a short time that it effectively stops in an instant? (This is called Zeno behavior, named after the ancient Greek philosopher Zeno and his paradoxes about motion).

Usually, mathematicians have to invent a different, complicated rule (called a "Lyapunov function") for every single type of stability. It's like having a different key for every single door in a giant castle.

The Big Idea of This Paper
The authors, Joe Moeller and Aaron D. Ames, say: "Stop making new keys for every door! Let's build a universal master key."

They use a branch of math called Category Theory (think of it as the "grammar" or "blueprint" of mathematics) to show that all these different stability rules are actually the same thing, just looking at the problem from different angles.

Here is how they do it, using some simple analogies:

1. The "Hybrid System" as a Storyteller

Imagine a hybrid system as a storyteller who tells a story with two types of sentences:

Continuous Sentences: "The ball is falling..." (Smooth flow).
Discrete Sentences: "The ball hit the ground and bounced!" (Sudden jump).

The authors realized that you can describe any of these systems using a single, unified mathematical structure. They call this structure a Coalgebra.

Analogy: Think of a coalgebra as a "machine blueprint." Instead of just drawing the machine, the blueprint tells you: "If you are in this state, here is how you flow, and here is where you might jump."

2. The "Lyapunov Morphism" (The Universal Translator)

To prove a system is stable, you usually need a "Lyapunov function." In simple terms, this is like a scorecard or a thermometer.

If the score goes down over time, the system is calming down (stable).
If the score goes up, the system is getting chaotic (unstable).

The authors discovered that a Lyapunov function is actually a morphism.

Analogy: Imagine you have a complex, messy machine (your hybrid system). You want to know if it's safe. Instead of analyzing the messy machine directly, you build a simple, perfect model of a stable machine (like a perfect, damped pendulum that always stops).
A Morphism is a translator that takes the messy machine and maps it onto the perfect model.
If the messy machine's "score" behaves exactly like the perfect model's score, then your messy machine is also stable!

The magic of this paper is that different types of stability (like "will it stop?" vs. "will it stop quickly?" vs. "will it stop in zero time?") are just different choices of the perfect model you map to.

Want to check if it stops eventually? Map it to a model that slowly fades to zero.
Want to check if it stops instantly (Zeno stability)? Map it to a model that collapses to zero in a flash.

3. The "Zeno" Problem (The Infinite Bounce)

The paper focuses heavily on Zeno Stability.

The Scenario: Imagine a ball bouncing. Every time it hits the ground, it loses a little energy. It bounces lower and lower.
The Paradox: Mathematically, it might bounce an infinite number of times, but the total time it takes to do all those bounces is finite. It stops bouncing in a split second. This is "Zeno behavior."
The New Tool: The authors created a specific "Zeno Model" (a measurement object). By mapping their complex robot or ball system to this Zeno Model, they can instantly prove: "Yes, this system will undergo infinite jumps in finite time and settle down safely."

They even derived a new "summability bound."

Analogy: Before, we knew the ball would stop. Now, this paper gives us a precise calculator to say: "The ball will bounce for exactly 2.5 seconds total, no matter how hard you throw it."

4. The "Simulation" Trick (Copying Stability)

One of the coolest parts of the paper is Simulation Morphisms.

The Problem: Analyzing a complex robot leg is hard. Analyzing a simple bouncing ball is easy.
The Solution: The authors show that if you can translate (simulate) the robot leg's behavior into the behavior of the bouncing ball, and you know the ball is stable, then the robot leg is automatically stable too!
Analogy: It's like testing a new, complex car engine by running it through a simulator that mimics a simple, reliable lawnmower engine. If the lawnmower engine never breaks, and your car engine behaves exactly like the lawnmower engine in the simulation, you know your car engine is safe.

Summary: Why Does This Matter?

Before this paper, engineers had to invent a new, complicated math proof for every new hybrid system they built.

Old Way: "Here is a walking robot. Let me invent Rule A. Now here is a jumping robot. Let me invent Rule B."
New Way (This Paper): "All these systems are just different versions of the same underlying structure. If we translate them into our 'Universal Stable Model,' we can use one single set of rules to prove they are all safe."

This unification allows engineers to:

Prove stability faster for complex robots (like bipedal walkers).
Predict Zeno behavior (infinite jumps in finite time) with precise mathematical bounds.
Reuse old proofs for new machines by simply "translating" the new machine into the language of an old, proven one.

In short, they took a messy, confusing pile of different stability rules and organized them into a single, elegant library where every book is just a different cover on the same story.

1. Problem Statement

Hybrid dynamical systems, which couple continuous flows with discrete jumps, are ubiquitous in robotics (e.g., bipedal locomotion, manipulation). A central challenge in this field is stability analysis. While Lyapunov methods are well-established for purely continuous or purely discrete systems, the theory for hybrid systems remains fragmented. Existing results for different stability notions (Lyapunov stability, asymptotic stability, exponential stability, and specifically Zeno stability) are treated as separate, ad-hoc theorems.

The authors identify a lack of unification: there is no single framework that explains why these different stability notions work or how they relate. Furthermore, analyzing Zeno behavior (infinite discrete transitions in finite time) and Zeno equilibria is particularly difficult because standard Lyapunov functions often fail to capture the convergence of the "Zeno time."

2. Methodology: Categorical Coalgebraic Framework

The authors employ Category Theory to unify these concepts. They model hybrid systems as coalgebras of an endofunctor $H$ on a specific category called Chart.

A. The Category Chart and Functor $H$

Chart: A category blending smooth manifolds (continuous dynamics) and sets (discrete modes). Objects are pairs $(M, S)$ , where $M$ is a manifold and $S$ is a set. Morphisms consist of a smooth map between manifolds and a function involving the set and manifold.
The Functor $H$ : Defined as $H(S, M) = (P(S \times M), TM)$ $H (S, M) = (P (S \times M), T M)$ .
- $TM$ (Tangent Bundle) captures continuous dynamics (vector fields).
- $P(S \times M)$ (Power Set) captures discrete dynamics (non-deterministic jumps and guards).
Hybrid Systems as Coalgebras: A hybrid system is an $H$ $H$ -coalgebra $( (S, M), (f_d, f_c) )$ $((S, M), (f_{d}, f_{c}))$ , where:
- $f_c: M \to TM$ defines the vector field.
- $f_d: S \times M \to P(S \times M)$ defines the guard and reset map (assigning a set of possible next states).
- This formulation generalizes classical hybrid automata, allowing for non-deterministic jumps and guards that are not necessarily submanifolds.

B. Categorical Lyapunov Theory

The paper instantiates a general categorical Lyapunov theorem (from prior work by the authors) to the $H$ -coalgebra setting.

Measurement Object ( $\sigma$ ): Stability is defined by mapping the system to a "simple target system" $\sigma$ . Different stability notions correspond to different choices of $\sigma$ .
Lyapunov Morphism: A Lyapunov function is redefined as a morphism from the system $(S, M)$ to the measurement object $\sigma$ .
Lax Commutativity: The core condition is that the diagram of the system and the target must "lax commute," meaning the derivative of the Lyapunov function along the system's flow is bounded by the dynamics of the target system (inequalities replace equalities).

3. Key Contributions

Unified Framework: The authors prove that Lyapunov stability, asymptotic stability, exponential stability, and Zeno stability are all instances of a single theorem: the existence of a morphism to a specific target system $\sigma$ .
New Zeno Stability Conditions: By instantiating the framework with a specific Zeno measurement object, they derive novel conditions for Zeno stability.
- This object forces the continuous component of the Lyapunov function to decrease linearly ( $\dot{V}_c = -c$ ) and the discrete component to contract geometrically ( $V_d^+ = \lambda V_d$ ).
- This yields a summability bound on the inter-jump durations, proving that the total time to reach the Zeno equilibrium is finite.
Simulation Morphisms and Transference: They introduce simulation morphisms between hybrid systems. This allows stability properties to be "transferred" from a simple system (like a bouncing ball) to a complex system (like a Lagrangian robot) via a mapping $\Phi$ .
Recovery of Classical Results: The framework naturally recovers known results for hybrid periodic orbits (including Rapid Exponential Stability) and switching systems.

4. Key Results

A. General Hybrid Lyapunov Theorem (Theorem 27)

For an $H$ -coalgebra to be stable with respect to a generalized element $z^*$ , there must exist a Lyapunov morphism $(V_d, V_c)$ satisfying:

Positive Definiteness: $V$ bounds the distance to $z^*$ .
Flow Condition: $\frac{dV_c}{dt} \cdot f_c(x) \leq \sigma_c(V_c(x))$ .
Jump Condition: The set of values after a jump is bounded by the target dynamics $\sigma_d$ .

B. Zeno Stability (Theorem 32)

Using the Zeno measurement object, the authors prove that if a hybrid system admits a Lyapunov morphism satisfying specific flow and jump inequalities, then:

The system is Zeno stable (converges to the Zeno equilibrium).
The Zeno time is finite. Specifically, the sum of inter-jump times is bounded by:
$\sum \Delta t_k \leq \frac{V_c(x_0)}{c} + \frac{V_d(x_0)}{c(1-\lambda)}$
This provides a concrete upper bound on the time until the Zeno behavior occurs.

C. Application to Lagrangian Systems (Example 39)

The authors apply the Transference Theorem to Lagrangian hybrid systems (e.g., robots with impacts).

Instead of constructing a complex Lyapunov function from scratch, they map the Lagrangian system to a Bouncing Ball system via a chart $\Phi$ .
They pull back the known Lyapunov function of the bouncing ball to the Lagrangian system.
Result: They prove Zeno stability for a broad class of Lagrangian systems with unilateral constraints, recovering pointwise stability in 1D and set-stability in higher dimensions.

5. Significance

Theoretical Unification: The paper resolves the fragmentation in hybrid stability theory by showing that diverse stability concepts are merely different instantiations of a single categorical structure.
Rigorous Zeno Analysis: It provides the first rigorous, constructive conditions for Zeno stability that include a bound on the Zeno time, a critical metric for control design in robotics.
Modularity via Simulation: The introduction of simulation morphisms allows engineers to analyze complex hybrid systems by reducing them to simpler, well-understood models (like the bouncing ball), significantly simplifying the verification process for robotic systems with impacts.
Categorical Power: It demonstrates the practical utility of category theory in control engineering, moving beyond abstract algebra to provide concrete, computable stability conditions for real-world hybrid systems.

In summary, Moeller and Ames provide a powerful, unified language for hybrid systems that not only recovers existing stability results but also generates new, stronger conditions for Zeno behavior, directly applicable to the stability analysis of advanced robotic systems.