Data-driven inversion-based control of nonlinear systems

By S. Formentin, Carlo Novara, S. Savaresi, Mario Milanese. 2015

2,416 citations recorded by the source corpus.

Abstract In this paper, we introduce the Data-Driven Inversion-Based Control (D2-IBC) method for nonlinear control system design. The method relies on a two degree-of-freedom architecture, with a nonlinear controller and a linear controller running in parallel, and does not require any detailed physical knowledge of the plant to control. Specifically, we use input/output data to synthesize the control action by employing convex optimization tools only. We show the effectiveness of the proposed approach on a simulation example, where the D2-IBC performance is also compared to that of the Direct FeedbacK (DFK) design approach, a benchmark method for nonlinear controller design from data.

Subfields: Control and Systems Engineering

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

Citations in the corpus

Mitigating Information Leakage in Image Representations: A Maximum Entropy Approach (2019, positive): “[16] both leveraged tools from non-linear systems theory [9] to analyze the convergence properties of simultaneous gradient descent in the context of GANs.”
A novel actor-critic-identifier architecture for approximate optimal control of uncertain nonlinear systems (2013, background): “…continuously differentiable and the Jacobian ∂Ω nom ∂ ˜ W c = −η c Γ ψψ T is bounded for the exponentially stable system in (44), the converse Lyapunov Theorem 4.14 in Khalil (2002) can be used to show that there exists a function V c : R N × [0, ∞) → R, which satisfies the following inequalities…”
Synchronization and transient stability in power networks and non-uniform Kuramoto oscillators (2009, methodology): “The analysis by Tikhonov’s method [24, 31] requires a system evolving on Euclidean space and an exponentially stable fixed point.”
Reaching a consensus via pinning control (2009, background): “However, virtually all physical systems are nonlinear in nature (Khalil, 2002).”
Mathematical Systems Theory I (2006, background): “For other introductory treatments to (mostly) nonlinear systems theory, the books [1], [4], [5] and [7] are all well regarded and certainly worthy of consideration.”
R&D Networks: Theory, Empirics, and Policy Implications (2019, background): “Because f(x) ≤ g(x), the comparison lemma implies that x(t) ≤ y(t) for all t ≥ 0 (see Khalil [2002], Lemma 3.”
Recent developments on the stability of systems with aperiodic sampling: An overview (2017, methodology): “The simplest approach consists in designing first a continuous-time controller using classical methods (Isidori, 1995; Khalil, 2002; Krstic, Kanellakopoulos, & Kokotovic, 1995; Sastry, 1999).”
Identification of block-oriented nonlinear systems using orthonormal bases (2004, background): “Although there are well established results concerning the stability of FBO models [24], less attention has been paid to the problem of identification of these model structures (see for instance [2]).”
Input–Output Finite-Time Stability of Linear Systems: Necessary and Sufficient Conditions (2009, background): “Roughly speaking (the precise definition is more involved, the interested reader is referred to [2], Ch.”
A Survey of Recent Results in Networked Control Systems (2007, methodology): “For deterministic systems we use standard notions of stability, e.g., from [21].”
Output-feedback adaptive stabilization control design for non-holonomic systems with strong non-linear drifts (2005, background): “This together with Barbălat’s Lemma (Khalil, 2002) implies that e and zi (i¼ 0, 1, . . . , n) are uniformly continuous, and further, together with (65), implies that lim t!”
Analysis and Design of Optimization Algorithms via Integral Quadratic Constraints (2014, background): “This formulation allows one to directly answer questions such as “what range of stepsizes can yield a given rate?”.”
Controller Synthesis for String Stability of Vehicle Platoons (2014, background): “It is noted that Definition 1 closely resembles the common L2 stability definition [16] as far as (nonstrict) L2 string stability is concerned, except for the fact that the inequality (10) must hold for all string lengths m ∈ N.”