arXiv subjects

Roland Toth

Publications and source records attributed to Roland Toth.

9 recordsLinked to original sources

arXiv2026

Efficient sparse GP-MPC with accurate mean and variance propagation applied for quadcopter flight control

This paper presents a computationally efficient approach for Gaussian process model predictive control (GP-MPC), where Gaussian process (GP) regression is used to complement a baseline model of the system dynamics. The proposed method achieves propagation of both the predicted mean and variance, thereby significantly reducing conservativeness compared with existing GP-MPC formulations. The nonlinear GP-MPC problem is reformulated into an exact linear parameter-varying (LPV) structure that preserves the nonlinear prediction dynamics in affine form without introducing further approximation. Moreover, closed-form derivations of moment matching (MM) predictions for sparse GPs are developed, including both mean and variance propagation under uncertain inputs, which improves scalability to larger datasets. This further enables recasting the resulting GP-MPC problem as a sequence of quadratic programs (QPs), which can be solved efficiently. The proposed framework significantly improves runtime efficiency while maintaining prediction accuracy, as demonstrated through simulation and real-world experiments on a Crazyflie 2.1 micro quadcopter.

math.OC↗

arXiv2023

Curve Trajectory Model for Human Preferred Path Planning of Automated Vehicles

Automated driving systems are often used for lane keeping tasks. By these systems, a local path is planned ahead of the vehicle. However, these paths are often found unnatural by human drivers. We propose a linear driver model, which can calculate node points that reflect the preferences of human drivers and based on these node points a human driver preferred motion path can be designed for autonomous driving. The model input is the road curvature. We apply this model to a self-developed Euler-curve-based curve fitting algorithm. Through a case study, we show that the model based planned path can reproduce the average behavior of human curve path selection. We analyze the performance of the proposed model through statistical analysis that shows the validity of the captured relations.

cs.RO↗

arXiv2021

Non-linear State-space Model Identification from Video Data using Deep Encoders

Identifying systems with high-dimensional inputs and outputs, such as systems measured by video streams, is a challenging problem with numerous applications in robotics, autonomous vehicles and medical imaging. In this paper, we propose a novel non-linear state-space identification method starting from high-dimensional input and output data. Multiple computational and conceptual advances are combined to handle the high-dimensional nature of the data. An encoder function, represented by a neural network, is introduced to learn a reconstructability map to estimate the model states from past inputs and outputs. This encoder function is jointly learned with the dynamics. Furthermore, multiple computational improvements, such as an improved reformulation of multiple shooting and batch optimization, are proposed to keep the computational time under control when dealing with high-dimensional and large datasets. We apply the proposed method to a video stream of a simulated environment of a controllable ball in a unit box. The study shows low simulation error with excellent long term prediction capability of the model obtained using the proposed method.

eess.SY↗

arXiv2021

Nonlinear state-space identification using deep encoder networks

Nonlinear state-space identification for dynamical systems is most often performed by minimizing the simulation error to reduce the effect of model errors. This optimization problem becomes computationally expensive for large datasets. Moreover, the problem is also strongly non-convex, often leading to sub-optimal parameter estimates. This paper introduces a method that approximates the simulation loss by splitting the data set into multiple independent sections similar to the multiple shooting method. This splitting operation allows for the use of stochastic gradient optimization methods which scale well with data set size and has a smoothing effect on the non-convex cost function. The main contribution of this paper is the introduction of an encoder function to estimate the initial state at the start of each section. The encoder function estimates the initial states using a feed-forward neural network starting from historical input and output samples. The efficiency and performance of the proposed state-space encoder method is illustrated on two well-known benchmarks where, for instance, the method achieves the lowest known simulation error on the Wiener--Hammerstein benchmark.

cs.LG↗

arXiv2020

Toolbox for Discovering Dynamic System Relations via TAG Guided Genetic Programming

Data-driven modeling of nonlinear dynamical systems often require an expert user to take critical decisions a priori to the identification procedure. Recently an automated strategy for data driven modeling of \textit{single-input single-output} (SISO) nonlinear dynamical systems based on \textit{Genetic Programming} (GP) and \textit{Tree Adjoining Grammars} (TAG) has been introduced. The current paper extends these latest findings by proposing a \textit{multi-input multi-output} (MIMO) TAG modeling framework for polynomial NARMAX models. Moreover we introduce a TAG identification toolbox in Matlab that provides implementation of the proposed methodology to solve multi-input multi-output identification problems under NARMAX noise assumption. The capabilities of the toolbox and the modelling methodology are demonstrated in the identification of two SISO and one MIMO nonlinear dynamical benchmark models.

eess.SY↗

arXiv2020

Linear Parameter-Varying Embedding of Nonlinear Models with Reduced Conservativeness

In this paper, a systematic approach is developed to embed the dynamical description of a nonlinear system into a linear parameter-varying (LPV) system representation. Initially, the nonlinear functions in the model representation are approximated using multivariate polynomial regression. Taking into account the residuals of the approximation as the potential scheduling parameters, a principle component analysis (PCA) is conducted to introduce a limited set of auxiliary scheduling parameters in coping with the trade-o? between model accuracy and complexity. In this way, LPV embedding of the nonlinear systems and scheduling variable selection are jointly performed such that a good trade-o? between complexity and conservativeness can be found. The developed LPV model depends polynomially on some of the state variables and affinely on the introduced auxiliary scheduling variables, which all together comprise the overall scheduling vector. The methodology is applied to a two-degree of freedom (2-DOf) robotic manipulator in addition to an academic example to reveal the effectiveness of the proposed method and to show the merits of the presented approach compared with some available results in the literature.

eess.SY↗

arXiv2020

Affine Linear Parameter-Varying Embedding of Nonlinear Models with Improved Accuracy and Minimal Overbounding

In this paper, automated generation of linear parameter-varying (LPV) state-space models to embed the dynamical behavior of nonlinear systems is considered, focusing on the trade-off between scheduling complexity and model accuracy and on the minimization of the conservativeness of the resulting embedding. The LPV state-space model is synthesized with affine scheduling dependency, while the scheduling variables themselves are nonlinear functions of the state and input variables of the original system. The method allows to generate complete or approximative embedding of the nonlinear system model and also it can be used to minimize complexity of existing LPV embeddings. The capabilities of the method are demonstrated on simulation examples and also in an empirical case study where the first-principle motion model of a 3-DOF control moment gyroscope is converted by the proposed method to LPV model with low scheduling complexity. Using the resulting model, a gain-scheduled controller is designed and applied on the gyroscope, demonstrating the efficiency of the developed approach.

eess.SY↗

arXiv2020

On the Initialization of Nonlinear LFR Model Identification with the Best Linear Approximation

Balancing the model complexity and the representation capability towards the process to be captured remains one of the main challenges in nonlinear system identification. One possibility to reduce model complexity is to impose structure on the model representation. To this end, this work considers the linear fractional representation framework. In a linear fractional representation the linear dynamics and the system nonlinearities are modeled by two separate blocks that are interconnected with one another. This results in a structured, yet flexible model structure. Estimating such a model directly from input-output data is not a trivial task as the involved optimization is nonlinear in nature. This paper proposes an initialization scheme for the model parameters based on the best linear approximation of the system and shows that this approach results in high quality models on a set of benchmark data sets.

eess.SY↗

arXiv2015

Moment Matching Based Model Reduction for LPV State-Space Models

We present a novel algorithm for reducing the state dimension, i.e. order, of linear parameter varying (LPV) discrete-time state-space (SS) models with affine dependence on the scheduling variable. The input-output behavior of the reduced order model approximates that of the original model. In fact, for input and scheduling sequences of a certain length, the input-output behaviors of the reduced and original model coincide. The proposed method can also be interpreted as a reachability and observability reduction (minimization) procedure for LPV-SS representations with affine dependence.

eess.SY↗