Physics–Guided Neural Networks for Feedforward Control: From Consistent Identification to Feedforward Controller Design

Bolderman, Max; Lazăr, Mircea; Butler, Hans

doi:10.1109/cdc51059.2022.9992852

Cited by 3 publications

(7 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This typically yields a two-step feedforward controller design procedure, consisting of an identification step to fit the model (5) to the data (3), and an inversion step to compute the feedforward input. Although this approach is generally adopted in literature, see, e.g., [3], [15], [16], [18], a quantitative relation between the identification and the tracking error, respectively, is still missing. Hence, the first problem considered in this work is to establish a quantitative relation between the tracking error on one hand, and the identification error and inversion error on the other hand.…”

Section: Problem Formulationmentioning

confidence: 99%

“…Often, the nonlinear inverse model-based feedforward controller suffers a performance loss when the forward model is nonminimum phase. For example, [3] imposes stability of PGNN feedforward controllers by constraining the NN parameters based on sufficient conditions for stability. This might unnecessarily limit the flexibility of the PGNN model, and, consequently, the accuracy of the identification.…”

Section: Problem Formulationmentioning

confidence: 99%

“…Remark 4.2: In the (PG)NN-based feedforward control literature it is common to employ a regularization term in the cost function [3,Section 4] and to use the squared 2norm due to its smoothness, such that…”

Section: Feedforward Control-orientedmentioning

confidence: 99%

“…Remark 4.3: A similar derivation as presented in this section can be performed for a direct closed-loop identification. This approach has the potential for consistent identification when noise is present in the data set, see, e.g., [21].…”

Section: Feedforward Control-orientedmentioning

confidence: 99%

“…Indeed, linear models are not capable to describe the complete dynamical behaviour of a mechatronic system, which includes unknown dynamics arising from, e.g., manufacturing tolerances [2]. Consequently, there is an interest to employ rich, nonlinear models for feedforward control that can learn the unknown dynamics from data [3]. This, however, brings (new) challenges to feedforward control design, which are briefly discussed next.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Data-driven feedforward control design for nonlinear systems: A control-oriented system identification approach

Bolderman¹,

Lazăr²,

Butler³

2023

Preprint

View full text Add to dashboard Cite

Feedforward controllers typically rely on accurately identified inverse models of the system dynamics to achieve high reference tracking performance. However, the impact of the (inverse) model identification error on the resulting tracking error is only analyzed a posteriori in experiments. Therefore, in this work, we develop an approach to feedforward control design that aims at minimizing the tracking error a priori. To achieve this, we show that the norm of the tracking error can be upperbounded by the sum of the inversion error and the identification error. This yields a two-step feedforward control design procedure that consists of a feedforward controloriented system identification, and a finite-horizon optimization to compute the feedforward control signal. The nonlinear feedforward control design method is validated using physicsguided neural networks on a nonlinear, nonminimum phase mechatronic example, where it outperforms standard ILC.

show abstract

Section: Problem Formulationmentioning

confidence: 99%

Section: Problem Formulationmentioning

confidence: 99%

Section: Feedforward Control-orientedmentioning

confidence: 99%

Section: Feedforward Control-orientedmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Data-driven feedforward control design for nonlinear systems: A control-oriented system identification approach

Bolderman¹,

Lazăr²,

Butler³

2023

Preprint

View full text Add to dashboard Cite

show abstract

Deep Learning Methods in Soft Robotics: Architectures and Applications

Čakurda,

Trojanová,

Pomin

et al. 2024

Advanced Intelligent Systems

View full text Add to dashboard Cite

The area of soft robotics has been subject to intense research efforts in the past two decades and constitutes a paradigm for advanced machine design in future robotic applications. However, standard methods for industrial robotics may be difficult to apply when analyzing soft robots. Deep learning, which has undergone rapid and transformative advancements in recent years, offers a set of powerful tools for analyzing and designing complex soft machines capable of operating in unstructured environments and interacting with humans and objects in a delicate manner. This review summarizes the most important state‐of‐the‐art deep learning architectures classified under supervised, unsupervised, semisupervised, and reinforcement learning scenarios and discusses their main features and benefits for different soft robotic applications, including soft robot manipulators, soft grippers, soft sensors, and e‐skins, as well as bioinspired soft robots. Specific properties of recent deep learning architectures and the usefulness of their features in addressing various types of issues found in soft robotics are analyzed. The existing challenges and future prospects are identified and discussed in view of the enhanced integration of both areas, which improves the performance of next‐generation soft machines operating in real‐world conditions.

show abstract

Data-Driven Feedforward Control Design for Nonlinear Systems: A Control-Oriented System Identification Approach

Bolderman,

Lazar,

Butler

2023

2023 62nd IEEE Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

Physics–Guided Neural Networks for Feedforward Control: From Consistent Identification to Feedforward Controller Design

Cited by 3 publications

References 21 publications

Data-driven feedforward control design for nonlinear systems: A control-oriented system identification approach

Data-driven feedforward control design for nonlinear systems: A control-oriented system identification approach

Deep Learning Methods in Soft Robotics: Architectures and Applications

Data-Driven Feedforward Control Design for Nonlinear Systems: A Control-Oriented System Identification Approach

Contact Info

Product

Resources

About