Skip to content

Free link for one month

February 14, 2018

IF you desire to freely access to the published paper Mixed ℓ2 and ℓ1-norm regularization for adaptive detrending with ARMA modeling and read it, you can use the following link that is active for 30 days




December 13, 2017

Mixed l2 and l1-norm regularization for adaptive detrending with ARMA modeling

L. Giarré and F. Argenti

To be published in Journal of the Franklin Institute

In this paper, the problem of detrending a time series and/or estimating a wandering baseline is addressed. We propose a new methodology that adaptively minimizes different regularized cost functions by introducing an ARMA model of the underlying trend. Mixed l_1 / l2-norm penalty functions are taken into consideration and novel RLS and LMS solutions are derived for the model parameters estimation. The proposed methods are applied to typical trend estimation/removal problems that can be found in the analysis of economic time series or biomedical signal acquisition. Comparisons with standard noncausal filtering techniques are also presented.

Maker Faire Eu @Rome 2017

December 3, 2017

With prof Federica Pascucci from university of Roma tre and Giovanni Galioto (ph.d. Student of the university of Palermo) and his advisor prof Ilenia Tinnirello and the effort of all the team of we spent the last three days at the Maker Faire in Rome. Our stand was located in the Robotic Aera ( organized by prof Bruno Siciliano) together with university of Pisa, Siena, Napoli, Bologna, ITT and Scuola sant’anna presenting most recents advanced results in the robotic field.

Arianna in the version that integrates cv algorithms with PDR was tested by thousands of people.

Big the appreciation ( na’ figata!!) and the compliments received by the many testers.

The experience for us was very appealing…

Students, researchers, common people, kids, couples from very young age ( the most skilled) to old scary grandfathers and nannies had the possibility of blindfolded walk along the path!!

Prof. Del Re seminar

September 28, 2017

Seminar Announcement

September 27Th, 2017

DIEF –  P2.1 class – 11am

Prof. L. Del RE

Director of the Institute for Design and Control of Mechatronical Systems

Johannes Kepler University, Linz, Austria.


Joint design of model and control for automotive control


Automotive controls usually are based on very complex maps-based structures (a modern Diesel Engine Control Unit includes up to 40.000 labels, which can be scalar variables or maps) and are very difficult to tune. Against this background, very powerful on-line optimization tools like CAMEO have been developed, which automatize but not avoid the tuning effort. In recent years, an increasing trend in direction of model based control with reduced tuning effort has appeared, which still does not encompass the majority of the control function, but potentially could change the whole picture. At the JKU in cooperation with several partners, methods have been developed to approach the problem from a different perspective. Instead of fixing a control structure and tuning the parameters by some optimization method, a combination of a recursive nonlinear system identification and a nonlinear receding horizon optimal control approach which can directly use the parameters of the model has been fixed. The nonlinear identification is based on polynomial NARX models, and uses a combination of DOE based input design, parameter estimation and pruning to produce. The control part is based on a modification of the C/GMRES algorithm by T. Ohtsuka, which has been rewritten for input/output, time discrete nonlinear system and can thus use directly the result of the estimation. The method has been applied to several engines, in this presentation experimental results on a BMW 2.0 L engine will be shown in comparison with the production values and alternatives with other methods. 2017

September 13, 2017


Cane 2.0

June 12, 2017

Let me cite two press articles recently appeared.

They present the prototype solution developed by Prof Rus at CSAIL mit and shown at ICRA in Singapore. I am a coauthor of the paper.

The work was partially supported by ABF.

The First is on The Economist
The second is on the MIT News

Wearable system @MIT

June 1, 2017

Today the group of Professor Rus will present at ICRA in Singapore the following result