Deep learning for model order reduction of multibody systems to minimal coordinates

Angeli, Andrea; Desmet, Wim; Naets, Frank

doi:10.1016/j.cma.2020.113517

Cited by 13 publications

(16 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The dynamic models obtained from the above symbolic multibody approach represent the key ingredient of our haptic device framework. Various competing-or even complementary-approaches can be used to reduce the computational complexity of the model, e.g., via deep learning techniques [31] or to make the most of object-oriented languages [32] and parallel programming [33]. Some results are promising but currently remain limited to rather simple MBS [31].…”

Section: Hardware Frameworkmentioning

confidence: 99%

“…Various competing-or even complementary-approaches can be used to reduce the computational complexity of the model, e.g., via deep learning techniques [31] or to make the most of object-oriented languages [32] and parallel programming [33]. Some results are promising but currently remain limited to rather simple MBS [31]. Regarding parallel computation, let us point out that symbolic generation lends itself perfectly to the vectorization of multibody models [8] and represents a promising avenue of exploitation of GPUs or FPGAs architectures in order to further improve model computation performances.…”

Section: Hardware Frameworkmentioning

confidence: 99%

“…Figure31. NTV demonstrator and its two main sensors: a rotating potentiometer and a binocular strain gauge for torque measurement.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Haptic Devices Based on Real-Time Dynamic Models of Multibody Systems

Docquier

Timmermans

Paul

2021

Sensors

View full text Add to dashboard Cite

Multibody modeling of mechanical systems can be applied to various applications. Human-in-the-loop interfaces represent a growing research field, for which increasingly more devices include a dynamic multibody model to emulate the system physics in real-time. In this scope, reliable and highly dynamic sensors, to both validate those models and to measure in real-time the physical system behavior, have become crucial. In this paper, a multibody modeling approach in relative coordinates is proposed, based on symbolic equations of the physical system. The model is running in a ROS environment, which interacts with sensors and actuators. Two real-time applications with haptic feedback are presented: a piano key and a car simulator. In the present work, several sensors are used to characterize and validate the multibody model, but also to measure the system kinematics and dynamics within the human-in-the-loop process, and to ultimately validate the haptic device behavior. Experimental results for both developed devices confirm the interest of an embedded multibody model to enhance the haptic feedback performances. Besides, model parameters variations during the experiments illustrate the infinite possibilities that such model-based configurable haptic devices can offer.

show abstract

Section: Hardware Frameworkmentioning

confidence: 99%

Section: Hardware Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

Haptic Devices Based on Real-Time Dynamic Models of Multibody Systems

Docquier

Timmermans

Paul

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In [14,15] an interesting approach based on the combination of deep learning and MB dynamics information was proposed to achieve this transformation. It allows reducing a generic MB model to minimal coordinates allowing the description of the EOMs through E-ODEs while not requiring a specific formulation or access to the constraint equations.…”

Section: Introductionmentioning

confidence: 99%

A Discrete-Time Extended Kalman Filter Approach Tailored for Multibody Models: State-Input Estimation

Adduci

Vermaut

Naets

et al. 2021

Sensors

Self Cite

View full text Add to dashboard Cite

Model-based force estimation is an emerging methodology in the mechatronic community given the possibility to exploit physically inspired high-fidelity models in tandem with ready-to-use cheap sensors. In this work, an inverse input load identification methodology is presented combining high-fidelity multibody models with a Kalman filter-based estimator and providing the means for an accurate and computationally efficient state-input estimation strategy. A particular challenge addressed in this work is the handling of the redundant state-description encountered in common multibody model descriptions. A novel linearization framework is proposed on the time-discretized equations in order to extract the required system model matrices for the Kalman filter. The presented framework is experimentally validated on a slider-crank mechanism. The nonlinear kinematics and dynamics are well represented through a rigid multibody model with lumped flexibilities to account for localized interaction phenomena among bodies. The proposed methodology is validated estimating the input torque delivered by a driver electro-motor together with the system states and comparing the experimental data with the estimated quantities. The results show the stability and accuracy of the estimation framework by only employing the angular motor velocity, measured by the motor encoder sensor and available in most of the commercial electro-motors.

show abstract

“…The novelty is that the autoencoder training includes the multibody physics information [3]. In this way, the autoencoder does not only perform a dimensionality reduction of the original coordinates but can be used for model order reduction.…”

mentioning

confidence: 99%

Deep learning of multibody minimal coordinates

Angeli

Naets

Desmet

2021

Proc Appl Math and Mech

Self Cite

View full text Add to dashboard Cite

• Multibody models typically include redundant coordinates and constraints leading to differential algebraic equations (DAEs). • An autoencoder neural network is trained to find the multibody minimal coordinates. • The obtained neural network function is used for the non-linear model order reduction eliminating the constraints and leading to ordinary differential equations (ODEs). • In order to avoid overfitting, the multibody physics information is embedded in the training. • This permits to combine the obtained reduced order model with common integration, control or estimation algorithms such as the Kalman filter.

show abstract

Deep learning for model order reduction of multibody systems to minimal coordinates

Abstract: ScienceDirect Comput. Methods Appl. Mech. Engrg. xxx (xxxx) xxx www.elsevier.com/locate/cma Highlights Deep learning for model order reduction of multibody systems to minimal coordinates Comput. Methods Appl. Mech. Engrg. xxx (xxxx) xxx

Cited by 13 publications

References 26 publications

Haptic Devices Based on Real-Time Dynamic Models of Multibody Systems

Haptic Devices Based on Real-Time Dynamic Models of Multibody Systems

A Discrete-Time Extended Kalman Filter Approach Tailored for Multibody Models: State-Input Estimation

Deep learning of multibody minimal coordinates

Contact Info

Product

Resources

About