1,720,982 research outputs found
Group-Based Dimensionality Reduction and Estimation for Heterogeneous Large-Scale Traffic Networks
No full textIdentification of nonlinear dynamical system with synthetic data: a preliminary investigation
No full textThis paper introduces a new rationale for learning nonlinear dynamical systems. The method makes use of an additional identification dataset, obtained without performing a new experiment on the system under study. The data are generated in an automatical manner, starting from a set of experimentally acquired measurements. In order to leverage the additional generated information, fundamental techniques from the machine learning field known as Semi-Supervised Learning (SSL) are employed and adapted. The problem is then cast as a regularized parametric learning problem. The effectiveness of the proposed approach is assessed on various nonlinear benchmark systems via repeated simulations, comparing the obtained results with a standard regularization method for learning parametric models. (C) 2018, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved
Kernel-based system identification with manifold regularization: A Bayesian perspective
No full textThis paper presents a nonparametric Bayesian interpretation of kernel-based function learning with manifold regularization. We show that manifold regularization corresponds to an additional likelihood term derived from noisy observations of the function gradient along the regressors graph. The hyperparameters of the method are estimated by a suitable empirical Bayes approach. The effectiveness of the method in the context of dynamical system identification is evaluated on a simulated linear system and on an experimental switching system setup. (C) 2022 Elsevier Ltd. All rights reserved
Sensory nerve conduction and nociception in the equine lower forelimb during perineural bupivacaine infusion along the palmar nerves
No full textThe purpose of this investigation was to study lateral palmar nerve (LPN) and medial palmar nerve (MPN) morphology and
determine nociception and sensory nerve conduction velocity (SNCV) following placement of continuous peripheral nerve block
(CPNB) catheters along LPN and MPN with subsequent bupivacaine (BUP) infusion. Myelinated nerve fiber distribution in
LPN and MPN was examined after harvesting nerve specimens in 3 anesthetized horses and processing them for morphometric
analysis. In 5 sedated horses, CPNB catheters were placed along each PN in both forelimbs. Horses then received in one forelimb
3 mL 0.125% BUP containing epinephrine 1:200 000 and 0.04% NaHCO3 per catheter site followed by 2 mL/h infusion over a 6-day
period, while in the other forelimb equal amounts of saline (SAL) solution were administered. The hoof withdrawal response
(HWR) threshold during pressure loading of the area above the dorsal coronary band was determined daily in both forelimbs.
On day 6 SNCV was measured under general anesthesia of horses in each limb’s LPN and MPN to detect nerve injury, followed by CPNB catheter removal. The SNCV was also recorded in 2 anesthetized non-instrumented horses (sham controls). In both
LPN and MPN myelinated fiber distributions were bimodal. The fraction of large fibers (. 7 mm) was greater in the MPN than
LPN (P , 0.05). Presence of CPNB catheters and SAL administration did neither affect measured HWR thresholds nor SNCVs,
whereas BUP infusion suppressed HWRs. In conclusion, CPNB with 0.125% BUP provides pronounced analgesia by inhibiting sensory nerve conduction in the distal equine forelimb
Kernel-based system identification with manifold regularization: A Bayesian perspective
No full textThis paper presents a nonparametric Bayesian interpretation of kernel-based function learning with manifold regularization. We show that manifold regularization corresponds to an additional likelihood term derived from noisy observations of the function gradient along the regressors graph. The hyperparameters of the method are estimated by a suitable empirical Bayes approach. The effectiveness of the method in the context of dynamical system identification is evaluated on a simulated linear system and on an experimental switching system setup
Traffic-light control in urban environment exploiting drivers’ reaction to the expected red lights duration
Full text linkTraffic congestion in urban environment is one of the most critical issue for drivers and city planners for both environment and efficiency reasons. Traffic lights are one of the main tools used to regulate traffic by diverting the drivers between different paths. Rational drivers, in turn, react to the traffic light duration by evaluating their options and, if necessary, by changing direction in order to reach their destination quicker. In this paper, we introduce a macroscopic traffic model for urban intersections that incorporates this rational behavior of the drivers. Then, we exploit it to show that, by providing additional information about the expected red-time duration to the drivers, one can decrease the amount of congestion in the network and the overall length of the queues at the intersections. Additionally, we develop a control policy for the traffic lights that exploits the reaction of the drivers in order to divert them to a different route to further increase the performances. These claims are supported by extensive numerical simulations
Kernel-based identification of asymptotically stable continuous-time linear dynamical systems
Full text linkIn many engineering applications, continuous-time models are preferred to discrete-time ones, in that they provide good physical insight and can be derived also from non-uniformly sampled data. However, for such models, model selection is a hard task if no prior physical knowledge is given. In this paper, we propose a non-parametric approach to infer a continuous-time linear model from data, by automatically selecting a proper structure of the transfer function and guaranteeing to preserve the system stability properties. By means of benchmark simulation examples, the proposed approach is shown to outperform state-of-the-art continuous-time methods, also in the critical case when short sequences of canonical input signals, like impulses or steps, are used for model learning
Nonparametric continuous-time identification of linear systems: theory, implementation and experimental results
No full textThis paper presents an algorithm for continuous-time identification of linear dynamical systems using kernel methods. When the system is asymptotically stable, also the identified model is guaranteed to share such a property. The approach embeds the selection of the model complexity through optimization of the marginal likelihood of the data thanks to its Bayesian interpretation. The output of the algorithm is the continuous-time transfer function of the estimated model. In this work, we show the algorithmic and computational details of the approach, and test it on real experimental data from an Electro Hydro-Static Actuator (EHSA)
- …
