Predictive models can be integrated in the sensing and monitoring methodologies of mechatronic systems in operation. When systems change or are subject to varying operating conditions, adaptivity of the models is needed. The goal of this paper is to enable this adaptivity by presenting a framework for continual learning. The framework aims to transfer and remember information from previously learned systems when a model is updated to new operating conditions. We achieve this by means of the following three key mechanisms. We first include physical information about the system, heavily regularizing the model output. Secondly, the usage of epistemic uncertainty, used as an indicator of the changing system, shows to what extend a transfer is desired. Last but not least the usage of a prior within a Bayesian framework allows to regularize models further according to previously obtained information. The last two principles are enabled thanks to the use of Bayesian neural networks. The methodology will be applied to a camfollower system in a simulation environment, where results show that previously trained systems are better remembered with an increase of 72% compared to normal training procedures.
<p>This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.</p> <p>Online monitoring of mechatronic systems and production environments helps operators to steer their processes into regions of optimal performance. Filtering methods, such as the Kalman filter, emerge in this setting thanks to their ability to process sensor data in real-time. Extensive engineering efforts are required to attain a behavioral model on the system dynamics for this filter. What is more, the model needs updating when facing varying conditions that are prevalent in mechatronic and production systems during operation. At the same time, datadriven methods excel in adapting models to match measurements without relying heavily on expert knowledge, but suffer from main drawbacks such as the lack of interpretability and the batchwise approach. This paper proposes a novel methodology relying on two key ideas. First a dual extended Kalman filter allows to jointly estimate and adapt model parameters in an online setting. Secondly, within this filter a hybrid model is embedded that depends on physics-based and data-driven parameters that are being updated, leading up to increased prediction capabilities and hence improved estimations on the state of the system. We evaluate the presented hybrid dual Extended Kalman filter on a cam follower system where it shows the ability to learn the cam shape during operation whilst tracking the state with an increased accuracy of more than 22% compared to other filtering techniques.</p>
<p>This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible.</p> <p>Online monitoring of mechatronic systems and production environments helps operators to steer their processes into regions of optimal performance. Filtering methods, such as the Kalman filter, emerge in this setting thanks to their ability to process sensor data in real-time. Extensive engineering efforts are required to attain a behavioral model on the system dynamics for this filter. What is more, the model needs updating when facing varying conditions that are prevalent in mechatronic and production systems during operation. At the same time, datadriven methods excel in adapting models to match measurements without relying heavily on expert knowledge, but suffer from main drawbacks such as the lack of interpretability and the batchwise approach. This paper proposes a novel methodology relying on two key ideas. First a dual extended Kalman filter allows to jointly estimate and adapt model parameters in an online setting. Secondly, within this filter a hybrid model is embedded that depends on physics-based and data-driven parameters that are being updated, leading up to increased prediction capabilities and hence improved estimations on the state of the system. We evaluate the presented hybrid dual Extended Kalman filter on a cam follower system where it shows the ability to learn the cam shape during operation whilst tracking the state with an increased accuracy of more than 22% compared to other filtering techniques.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.