Asymptotic stability is the property of an equilibrium where nearby solutions stay close and actually converge to that equilibrium as time goes on. In Linear Algebra and Differential Equations, you check it by looking at eigenvalues, linear systems, and sometimes Lyapunov functions.
Asymptotic stability is the stronger version of stability for an equilibrium point in a differential equation system. If you start the system with a small disturbance, the solution not only stays near the equilibrium, it also moves back toward it as t goes to infinity.
That extra word, asymptotic, is doing real work. A stable system can stay close without ever settling back exactly. An asymptotically stable system has a built-in pull toward the equilibrium, so the long-term behavior is predictable, not just bounded.
For linear systems, this idea often shows up through eigenvalues. If the system matrix has eigenvalues with negative real parts, solutions decay toward the equilibrium, usually the origin after you shift the system. That is why eigenvalues matter so much in this course, they tell you whether the flow points inward or outward over time.
You also see asymptotic stability when a system is nonhomogeneous. The forcing term can shift where the equilibrium sits or create a particular solution that the system settles around. In that setting, you still look at the homogeneous part first, because its long-term behavior tells you whether the system’s natural motion dies out or keeps oscillating.
A good way to picture it is a ball in a bowl. If you nudge it, a stable bowl keeps it from rolling away. An asymptotically stable bowl makes it roll back to the bottom. In contrast, if the bottom is flat or the bowl is tipped the wrong way, the motion may hover, drift, or blow up instead.
A common mistake is mixing up stability with asymptotic stability. Stable means nearby trajectories remain nearby. Asymptotically stable means they actually converge to the equilibrium. That distinction matters any time you are sketching phase portraits, classifying critical points, or checking whether a model settles into steady behavior.
Asymptotic stability is one of the main tools for predicting what a differential equation system does after time passes. In a linear algebra and differential equations course, you use it to decide whether a solution settles down, keeps oscillating, or runs away from equilibrium.
This comes up when you analyze systems with matrices, especially after diagonalizing or finding eigenvalues. The sign of the real part tells you whether the system’s natural motion shrinks toward the equilibrium or grows away from it. That turns abstract matrix work into a long-term behavior check.
It also matters in nonhomogeneous systems, where forcing terms add extra movement. You often separate the homogeneous part from the particular solution, then ask whether the forcing just shifts the steady state or whether it changes the stability picture. That is a big part of Topic 10.3 and a common problem set skill.
In applied settings, asymptotic stability is how you decide whether a control system stays on target after a disturbance. In class, that shows up as phase portraits, matrix calculations, or a Lyapunov function that keeps decreasing. If you can recognize asymptotic stability, you can explain not just what the solutions do now, but what they do in the long run.
Keep studying Linear Algebra and Differential Equations Unit 10
Visual cheatsheet
view galleryEquilibrium Point
Asymptotic stability is always measured relative to an equilibrium point. First you find the equilibrium, then you ask whether nearby solutions return to it or drift away. In many problems, the equilibrium is the origin after you rewrite the system, but for nonhomogeneous systems the equilibrium may shift to a different steady state.
Lyapunov Stability
Lyapunov stability is the weaker idea: nearby solutions stay nearby. Asymptotic stability adds convergence back to equilibrium. That difference is easy to miss on a quiz, so check whether the problem asks for boundedness near the point or actual return to the point over time.
Nonhomogeneous Systems
A nonhomogeneous system includes a forcing term, so the long-term behavior is not just the homogeneous solution by itself. You often solve for a particular solution first, then study whether the homogeneous part decays. Asymptotic stability helps you decide if the full solution settles into a steady response.
Routh-Hurwitz Criterion
When you do not want to compute every eigenvalue directly, the Routh-Hurwitz Criterion can tell you whether the real parts are negative. That makes it a shortcut for checking asymptotic stability in certain linear systems and characteristic polynomials.
A problem set or quiz question will usually give you a system matrix, a characteristic equation, or a phase portrait and ask whether the equilibrium is asymptotically stable. Your job is to check the sign of the eigenvalues’ real parts, or use a stability test like Routh-Hurwitz when that is allowed. If the system is nonhomogeneous, separate the forcing from the homogeneous behavior and decide whether the solution settles toward a steady state.
You may also be asked to explain the difference between stable and asymptotically stable in words. A strong answer says that stable solutions stay near the equilibrium, while asymptotically stable ones actually converge to it as time increases. On sketch-based problems, look for inward flow, spirals that shrink, or trajectories that approach a fixed point.
These terms are closely related, but they are not the same. Lyapunov stability only guarantees that solutions starting near an equilibrium remain near it. Asymptotic stability adds the stronger condition that those solutions move back to the equilibrium over time. If a question asks whether trajectories just stay close or actually converge, that tells you which term to use.
Asymptotic stability means nearby solutions stay close to an equilibrium and also converge to it as time goes to infinity.
For linear systems, negative real parts of the eigenvalues usually signal asymptotic stability.
Stable is weaker than asymptotically stable, so do not treat those words as synonyms.
In nonhomogeneous systems, the forcing term can shift the steady behavior, but the homogeneous part still tells you a lot about long-term stability.
If you can read a phase portrait or characteristic equation, you can usually decide whether the system settles down or moves away.
It is the property that solutions starting near an equilibrium stay near it and then move back to it as time goes on. In this course, you usually check it with eigenvalues, phase portraits, or a Lyapunov argument. The long-term behavior matters as much as the short-term behavior.
Stability means nearby solutions do not drift far away from the equilibrium. Asymptotic stability means they do that and also converge to the equilibrium over time. So every asymptotically stable equilibrium is stable, but not every stable equilibrium is asymptotically stable.
For a linear system, if every eigenvalue has a negative real part, solutions decay toward the equilibrium. If any eigenvalue has a positive real part, the system is unstable. If real parts are zero, you usually need more information before deciding.
You usually see trajectories moving inward toward a fixed point. They may spiral in or approach straight in, depending on the eigenvalues, but the main sign is that the arrows point toward the equilibrium over time. If they circle without shrinking, that is not asymptotic stability.