-
On identifying the non-linear dynamics of a hovercraft using an end-to-end deep learning approach
Authors:
Roland Schwan,
Nicolaj Schmid,
Etienne Chassaing,
Karim Samaha,
Colin N. Jones
Abstract:
We present the identification of the non-linear dynamics of a novel hovercraft design, employing end-to-end deep learning techniques. Our experimental setup consists of a hovercraft propelled by racing drone propellers mounted on a lightweight foam base, allowing it to float and be controlled freely on an air hockey table. We learn parametrized physics-inspired non-linear models directly from data…
▽ More
We present the identification of the non-linear dynamics of a novel hovercraft design, employing end-to-end deep learning techniques. Our experimental setup consists of a hovercraft propelled by racing drone propellers mounted on a lightweight foam base, allowing it to float and be controlled freely on an air hockey table. We learn parametrized physics-inspired non-linear models directly from data trajectories, leveraging gradient-based optimization techniques prevalent in machine learning research. The chosen model structure allows us to control the position of the hovercraft precisely on the air hockey table. We then analyze the prediction performance and demonstrate the closed-loop control performance on the real system.
△ Less
Submitted 15 May, 2024;
originally announced May 2024.
-
Physics-Informed Machine Learning for Modeling and Control of Dynamical Systems
Authors:
Truong X. Nghiem,
Ján Drgoňa,
Colin Jones,
Zoltan Nagy,
Roland Schwan,
Biswadip Dey,
Ankush Chakrabarty,
Stefano Di Cairano,
Joel A. Paulson,
Andrea Carron,
Melanie N. Zeilinger,
Wenceslao Shaw Cortez,
Draguna L. Vrabie
Abstract:
Physics-informed machine learning (PIML) is a set of methods and tools that systematically integrate machine learning (ML) algorithms with physical constraints and abstract mathematical models developed in scientific and engineering domains. As opposed to purely data-driven methods, PIML models can be trained from additional information obtained by enforcing physical laws such as energy and mass c…
▽ More
Physics-informed machine learning (PIML) is a set of methods and tools that systematically integrate machine learning (ML) algorithms with physical constraints and abstract mathematical models developed in scientific and engineering domains. As opposed to purely data-driven methods, PIML models can be trained from additional information obtained by enforcing physical laws such as energy and mass conservation. More broadly, PIML models can include abstract properties and conditions such as stability, convexity, or invariance. The basic premise of PIML is that the integration of ML and physics can yield more effective, physically consistent, and data-efficient models. This paper aims to provide a tutorial-like overview of the recent advances in PIML for dynamical system modeling and control. Specifically, the paper covers an overview of the theory, fundamental concepts and methods, tools, and applications on topics of: 1) physics-informed learning for system identification; 2) physics-informed learning for control; 3) analysis and verification of PIML models; and 4) physics-informed digital twins. The paper is concluded with a perspective on open challenges and future research opportunities.
△ Less
Submitted 24 June, 2023;
originally announced June 2023.
-
PIQP: A Proximal Interior-Point Quadratic Programming Solver
Authors:
Roland Schwan,
Yuning Jiang,
Daniel Kuhn,
Colin N. Jones
Abstract:
This paper presents PIQP, a high-performance toolkit for solving generic sparse quadratic programs (QP). Combining an infeasible Interior Point Method (IPM) with the Proximal Method of Multipliers (PMM), the algorithm can handle ill-conditioned convex QP problems without the need for linear independence of the constraints. The open-source implementation is written in C++ with interfaces to C, Pyth…
▽ More
This paper presents PIQP, a high-performance toolkit for solving generic sparse quadratic programs (QP). Combining an infeasible Interior Point Method (IPM) with the Proximal Method of Multipliers (PMM), the algorithm can handle ill-conditioned convex QP problems without the need for linear independence of the constraints. The open-source implementation is written in C++ with interfaces to C, Python, Matlab, and R leveraging the Eigen3 library. The method uses a pivoting-free factorization routine and allocation-free updates of the problem data, making the solver suitable for embedded applications. The solver is evaluated on the Maros-Mészáros problem set and optimal control problems, demonstrating state-of-the-art performance for both small and large-scale problems, outperforming commercial and open-source solvers.
△ Less
Submitted 15 September, 2023; v1 submitted 1 April, 2023;
originally announced April 2023.
-
On continuation and convex Lyapunov functions
Authors:
Wouter Jongeneel,
Roland Schwan
Abstract:
Suppose that the origin is globally asymptotically stable under a set of continuous vector fields on Euclidean space and suppose that all those vector fields come equipped with -- possibly different -- convex Lyapunov functions. We show that this implies there is a homotopy between any two of those vector fields such that the origin remains globally asymptotically stable along the homotopy. Relaxi…
▽ More
Suppose that the origin is globally asymptotically stable under a set of continuous vector fields on Euclidean space and suppose that all those vector fields come equipped with -- possibly different -- convex Lyapunov functions. We show that this implies there is a homotopy between any two of those vector fields such that the origin remains globally asymptotically stable along the homotopy. Relaxing the assumption on the origin to any compact convex set or relaxing convexity to geodesic convexity does not alter the conclusion. Imposing the same convexity assumptions on control Lyapunov functions leads to a Hautus-like stabilizability test. These results ought to be of interest in the context of learning stability certificates, policy gradient methods and switched systems.
△ Less
Submitted 21 March, 2024; v1 submitted 14 January, 2023;
originally announced January 2023.
-
Violent suicide attempt history in elderly patients with bipolar disorder: The role of sex, abdominal obesity, and verbal memory: Results from the FACE-BD cohort (FondaMental Advanced center of Expertise for Bipolar Disorders)
Authors:
Aiste Lengvenyte,
Bruno Aouizerate,
Valerie Aubin,
Joséphine Loftus,
Emeline Marlinge,
Raoul Belzeaux,
Caroline Dubertret,
Sebastien Gard,
Emmanuel Haffen,
Raymund Schwan,
Pierre-Michel Llorca,
Christine Passerieux,
Paul Roux,
Mircea Polosan,
Bruno Etain,
Marion Leboyer,
Philippe Courtet,
Emilie Olié
Abstract:
Background: Bipolar disorder (BD) is a chronic, lifelong condition, associated with increased risk of obesity, cognitive impairment, and suicidal behaviors. Abdominal obesity and a higher risk of violent suicide attempt (SA) seem to be shared correlates with older age, BD, and male sex until middle age when menopause-related female body changes occur. This study aimed at assessing the role of abdo…
▽ More
Background: Bipolar disorder (BD) is a chronic, lifelong condition, associated with increased risk of obesity, cognitive impairment, and suicidal behaviors. Abdominal obesity and a higher risk of violent suicide attempt (SA) seem to be shared correlates with older age, BD, and male sex until middle age when menopause-related female body changes occur. This study aimed at assessing the role of abdominal obesity and cognition in the violent SA burden of individuals with BD.Methods: From the well-defined nationwide cohort FACE-BD (FondaMental Advanced center of Expertise for Bipolar Disorders), we extracted data on 619 euthymic BD patients that were 50 years or older at inclusion. Cross-sectional clinical, cognitive, and metabolic assessments were performed. SA history was based on self-report.Results: Violent SA, in contrast to non-violent and no SA, was associated with higher waist circumference, abdominal obesity and poorer California Verbal Learning Test short-delay free recall (CVLT-SDFR) (ANOVA, p < .001, p = .014, and p = .006). Waist circumference and abdominal obesity were associated with violent SA history independently of sex, BD type and anxiety disorder (Exp(B) 1.02, CI 1.00-1.05, p = .018; Exp(B) 2.16, CI 1.00-4.64, p = .009, accordingly). In an exploratory model, waist circumference and CVLT-SDFR performance mediated the association between male sex and violent SA.Limitations: Cross-sectional design and retrospective reporting.Conclusions: Violent SA history was associated with abdominal obesity and poorer verbal memory in older age BD patients. These factors were interlinked and might mediate the association between male sex and violent SA.
△ Less
Submitted 26 August, 2022;
originally announced August 2022.
-
Stability Verification of Neural Network Controllers using Mixed-Integer Programming
Authors:
Roland Schwan,
Colin N. Jones,
Daniel Kuhn
Abstract:
We propose a framework for the stability verification of Mixed-Integer Linear Programming (MILP) representable control policies. This framework compares a fixed candidate policy, which admits an efficient parameterization and can be evaluated at a low computational cost, against a fixed baseline policy, which is known to be stable but expensive to evaluate. We provide sufficient conditions for the…
▽ More
We propose a framework for the stability verification of Mixed-Integer Linear Programming (MILP) representable control policies. This framework compares a fixed candidate policy, which admits an efficient parameterization and can be evaluated at a low computational cost, against a fixed baseline policy, which is known to be stable but expensive to evaluate. We provide sufficient conditions for the closed-loop stability of the candidate policy in terms of the worst-case approximation error with respect to the baseline policy, and we show that these conditions can be checked by solving a Mixed-Integer Quadratic Program (MIQP). Additionally, we demonstrate that an outer and inner approximation of the stability region of the candidate policy can be computed by solving an MILP. The proposed framework is sufficiently general to accommodate a broad range of candidate policies including ReLU Neural Networks (NNs), optimal solution maps of parametric quadratic programs, and Model Predictive Control (MPC) policies. We also present an open-source toolbox in Python based on the proposed framework, which allows for the easy verification of custom NN architectures and MPC formulations. We showcase the flexibility and reliability of our framework in the context of a DC-DC power converter case study and investigate its computational complexity.
△ Less
Submitted 31 May, 2023; v1 submitted 27 June, 2022;
originally announced June 2022.
