Further Constructions of Control-Lyapunov Functions and Stabilizing Feedbacks for Systems Satisfying the Jurdjevic-Quinn Conditions

Frederic Mazenc,Michael Malisoff

DOI: https://doi.org/10.1109/TAC.2005.863500

2005-10-18

Abstract:For a broad class of nonlinear systems, we construct smooth control-Lyapunov functions whose derivatives along the trajectories of the systems can be made negative definite by smooth control laws that are arbitrarily small in norm. We assume our systems satisfy appropriate generalizations of the Jurdjevic-Quinn conditions. We also design state feedbacks of arbitrarily small norm that render our systems integral-input-to-state stable to actuator errors.

Optimization and Control

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to construct control Lyapunov functions (CLF) for a large class of nonlinear systems. The derivatives of the trajectories of these systems along the CLF can be made negative definite by smooth control laws. Specifically, the authors hope to design CLF and stable feedback that satisfy the following conditions:

1. **Applicable to a wide range of nonlinear systems**: The paper focuses on nonlinear systems that satisfy the generalized Jurdjevic - Quinn conditions.
2. **Explicitly expressed CLF**: Although it is known that for some systems, functions that make the derivative negative semi - definite can be constructed, explicit CLF expressions are usually difficult to obtain. Therefore, the authors' goal is to provide explicit CLF for these systems.
3. **Feedback control with arbitrarily small norms**: Design state feedback such that its norm is arbitrarily small and can make the system have integral input - to - state stability (iISS) with respect to actuator errors.

### Specific problem description

- **Constructing CLF**: For nonlinear systems that satisfy the generalized Jurdjevic - Quinn conditions, construct a smooth control Lyapunov function \( V^{\sharp}(x) \), and the derivative of it along the system trajectory can be made negative definite by the smooth control law \( u = -\xi(V(x))L_{g}V(x)^{T} \).

- **Stable feedback design**: Design state feedback \( u \) such that its norm is arbitrarily small and can make the system have integral input - to - state stability (iISS) with respect to actuator errors.

### Mathematical representation

Assume that the system has the following form:
\[ \dot{x}=F(x, u) \]
where \( x\in\mathbb{R}^{n} \) is the state, \( u\in\mathbb{R}^{m} \) is the control input, \( F(0, 0) = 0 \), and \( F \) is a \( C^{1} \) function.

The main results include:

1. **The form of CLF**:
\[ V^{\sharp}(x)=V(x)+\int_{0}^{V(x)}\Omega(s)ds+\delta(V(x))L_{g}V(x) \]
where \( V(x) \) is a known radially unbounded positive definite function, satisfying \( L_{f}V(x)\leq0 \).

2. **Feedback control**:
\[ u = -\xi(V(x))L_{g}V(x)^{T} \]
where \( \xi \) is a \( C^{1} \) positive function such that \( \xi(s)\leq\bar{\xi}(s) \) for all \( s\geq0 \).

Through these constructions, the authors prove that \( V^{\sharp}(x) \) is a CLF that satisfies the small - control property, and in the closed - loop system, its derivative is negative definite.

### Application example

The paper also takes a two - link manipulator as an example to show how to apply the above theory to construct CLF and stable feedback, and verifies the performance of these feedbacks.

### Summary

This paper solves the problem of constructing explicit CLF for nonlinear systems that satisfy the generalized Jurdjevic - Quinn conditions, and designs state feedback with arbitrarily small norms, making the system have integral input - to - state stability with respect to actuator errors. This result is of great significance for the design and analysis of nonlinear control systems.