Abstract:This paper presents a linear-programming based algorithm to perform data-driven stabilizing control of linear positive systems. A set of state-input-transition observations is collected up to magnitude-bounded noise. A state feedback controller and dual linear copositive Lyapunov function are created such that the set of all data-consistent plants is contained within the set of all stabilized systems. This containment is certified through the use of the Extended Farkas Lemma and solved via Linear Programming. Sign patterns and sparsity structure for the controller may be imposed using linear constraints. The complexity of this algorithm scales in a polynomial manner with the number of states and inputs. Effectiveness is demonstrated on example systems.

What problem does this paper attempt to address?

The problem this paper attempts to solve is: how to achieve stable control of positive linear systems through a data-driven approach using Linear Programming (LP) algorithms. Specifically, the paper focuses on designing a state feedback controller given a set of state-input-transition observation data, ensuring that all systems consistent with the observation data can be stabilized. Additionally, the paper extends this method to address the positive stability control problem of switched linear systems and Linear Parameter Varying (LPV) systems. ### Main Contributions of the Paper: 1. **Data-Driven Positive Stability Control**: Proposes an algorithm based on linear programming for stabilizing all positive linear systems consistent with the observation data. 2. **Computational Complexity Analysis**: Provides a detailed analysis of the computational complexity required to solve this linear programming problem. 3. **Worst-Case Peak-to-Peak Gain Minimization**: Extends the method to minimize the worst-case peak-to-peak gain from external input to controlled output. 4. **Application to Switched Positive Linear Systems**: Applies the method to the stability control of switched positive linear systems. 5. **Application to LPV Systems**: Further extends the method to handle the positive stability control of linear parameter varying systems. ### Method to Solve the Problem: - **Data Consistency Polytope**: Constructs a data consistency polytope to describe the set of system matrices consistent with the observation data. - **Stabilizing Control Polytope**: Constructs a stabilizing control polytope to describe the set of system matrices that can be stabilized by a specific controller. - **Extended Farkas' Lemma**: Utilizes the extended Farkas' lemma to ensure that the data consistency polytope is contained within the stabilizing control polytope, thereby guaranteeing that all systems consistent with the data can be stabilized. - **Linear Programming Solution**: Transforms the above problem into a linear programming problem and finds a suitable controller by solving this problem. ### Application Examples: - **Stability Control of Continuous-Time Systems**: Validates the effectiveness of the method through numerical experiments, demonstrating the stable trajectories of 100 randomly sampled systems under the controller's action. - **Stability Control of Discrete-Time Systems**: Similarly, validates the method's effectiveness in discrete-time systems through numerical experiments. ### Conclusion: The proposed method effectively achieves data-driven stable control of positive linear systems with low computational complexity. This method is not only applicable to individual systems but can also be extended to switched systems and linear parameter varying systems, showing broad application prospects.

Data-Driven Control of Positive Linear Systems using Linear Programming

Data-Driven Method for Response Control of Nonlinear Random Dynamical Systems

Data-driven Stabilization of SISO Feedback Linearizable Systems

Data-driven stabilization of switched and constrained linear systems

Data-Driven Optimal Control Using Perron-Frobenius Operator

Formulas for Data-driven Control: Stabilization, Optimality and Robustness

Feedback Control for Switched Positive Linear Systems

Scalable Control Design for K-positive Linear Systems

Data-driven input-to-state stabilization

A Linear Parameter-Varying Approach to Data Predictive Control

Data-Driven Structured Robust Control of Linear Systems

Direct Data-Driven State-Feedback Control of Linear Parameter-Varying Systems

Robust Control of Unknown Switched Linear Systems from Noisy Data

Data-driven optimal control via linear programming: boundedness guarantees

Linear programming-based robust model predictive control for positive systems

Data-based Transfer Stabilization in Linear Systems

Data-Based Control of Continuous-Time Linear Systems with Performance Specifications

Data-Driven Optimal Control Using Transfer Operators

Data-driven Control of Unknown Linear Systems via Quantized Feedback

A hybrid systems framework for data-based adaptive control of linear time-varying systems

Distributed Control of Positive Systems