D. Dalle Pezze, D. Deronjic, C. Masiero, D. Tosato, A. Beghi, G.A. Susto.
A Multi-label Continual Learning Framework to Scale Deep Learning Approaches for Packaging Equipment Monitoring. Engineering Applications of Artificial Intelligence, vol. 124, 2023 [
BibTeX]
A. Fabris, F. Giachelle, A. Piva, G. Silvello, G.A. Susto.
A Search Engine for Algorithmic Fairness Datasets. 2nd European Workshop on Algorithmic Fairness, 2023 [
BibTeX]
D. Dandolo, C. Masiero, M. Carletti, D. Dalle Pezze, G.A. Susto.
AcME - Accelerated Model-agnostic Explanations: Fast Whitening of the Machine-Learning Black Box. Expert Systems with Applications, vol. 214, 2023 [
url] [
BibTeX]
M. Fanan, C. Baron, R. Carli, M. Divernois, J. Marongiu, G.A. Susto.
Anomaly Detection for Hydroelectric Power Plants: a Machine Learning-based Approach. IEEE International Conference on Industrial Informatics (INDIN), 2023 [
BibTeX]
Q. Wang, T. Barbariol, G.A. Susto, B. Bonato, S. Guerra, U. Castiello.
Classifying Circumnutation in Pea Plants via Supervised Machine Learning. Plants, vol. 4(12), 2023 [
url] [
BibTeX]
F. Dalla Zuanna, N. Gentner, G.A. Susto.
Deep Learning-based Sequence Modeling for Advanced Process Control in Semiconductor Manufacturing. IFAC World Congress, 2023 [
BibTeX]
S. Toigo, A. Cenedese, D. Fornasier, B. Kasi.
Deep-learning based industrial quality control on low-cost smart cameras. Proc. SPIE 12749 - 16th International Conference on Quality Control by Artificial Vision (QCAV23), 2023
Abstract:
This paper aims to describe a combined machine vision and deep learning method for quality control in an
industrial environment. The innovative approach used for the proposed solution leverages the use of low-cost
hardware of reduced size, and yields extremely high evaluation accuracy and limited computational time. As a
result, the developed system works entirely on a portable smart camera. It does not require additional sensors,
such as photocells, nor is it based on external computation.
[
url] [
BibTeX]
D. Marcato, L. Bellan, D. Bortolato, M. Comunian, F. Gelain, V. Martinelli, G. Savarese, G.A. Susto.
Demonstration of Beam Emittance Optimization using Reinforcement Learning. 14th International Particle Accelerator Conference, 2023 [
BibTeX]
A. Fabris, G. Silvello, G.A. Susto, A. Biega.
Dissatisfaction Induced by Pairwise Swaps. Italian Information Retrieval Workshop, 2023 [
BibTeX]
L.C. Brito, G.A. Susto, J.N. Brito, M.A.V. Duarte.
Fault Diagnosis using eXplainable AI: a Transfer Learning-based Approach for Rotating Machinery exploiting Augmented Synthetic Data. Expert Systems with Applications, 2023 [
BibTeX]
M. Carletti, M. Terzi, G.A. Susto.
Interpretable Anomaly Detection with DIFFI: Depth-based Feature Importance for the Isolation Forest. Engineering Applications of Artificial Intelligence, vol. 119, 2023
Abstract:
Anomaly Detection is an unsupervised learning task aimed at detecting anomalous behaviors with respect to historical data. In particular, multivariate Anomaly Detection has an important role in many applications thanks to the capability of summarizing the status of a complex system or observed phenomenon with a single indicator (typically called ‘anomaly score’) and thanks to the unsupervised nature of the task that does not require human tagging. The Isolation Forest is one of the most commonly adopted algorithms in the field of Anomaly Detection due to its proven effectiveness and low computational complexity. A major problem affecting Isolation Forest is represented by the lack of interpretability, an effect of the inherent randomness governing the splits performed by the Isolation Trees, the building blocks of the Isolation Forest. In this paper, we propose effective yet computationally inexpensive methods to define feature importance scores at both global and local levels for the Isolation Forest. Moreover, we define a procedure to perform unsupervised feature selection for Anomaly Detection problems based on our interpretability method. Such a procedure also serves the purpose of tackling the challenging task of feature importance evaluation in unsupervised anomaly detection. We assess the performance on several synthetic and real-world datasets, including comparisons against state-of-the-art interpretability techniques, and make the code publicly available to enhance reproducibility and foster research in the field.
[
url] [
BibTeX]
S. McLoone, K. Guelton, T. Guerra, G.A. Susto, J. Kocijan, D. Romeres.
Introduction to the special issue on Intelligent Control and Optimisation. Engineering Applications of Artificial Intelligence, vol. 123, 2023 [
url] [
BibTeX]
F. Zocco, M. Maggipinto, G.A. Susto, S. McLoone.
Lazy FSCA for Unsupervised Variable Selection. Engineering Applications of Artificial Intelligence, vol. 124, 2023
Abstract:
Dimensionality reduction is a important step in the development of scalable and interpretable data-driven models, especially when there are a large number of candidate variables. This paper focuses on unsupervised variable selection based dimensionality reduction, and in particular on unsupervised greedy selection methods, which have been proposed by various researchers as computationally tractable approximations to optimal subset selection. These methods are largely distinguished from each other by the selection criterion adopted, which include squared correlation, variance explained, mutual information and frame potential. Motivated by the absence in the literature of a systematic comparison of these different methods, we present a critical evaluation of seven unsupervised greedy variable selection algorithms considering both simulated and real world case studies. We also review the theoretical results that provide performance guarantees and enable efficient implementations for certain classes of greedy selection function, related to the concept of submodularity. Furthermore, we introduce and evaluate for the first time, a lazy implementation of the variance explained based forward selection component analysis (FSCA) algorithm. Our experimental results show that: (1) variance explained and mutual information based selection methods yield smaller approximation errors than frame potential; (2) the lazy FSCA implementation has similar performance to FSCA, while being an order of magnitude faster to compute, making it the algorithm of choice for unsupervised variable selection.
[
url] [
BibTeX]
B. Pozzan, G. Giacomuzzo, M. Bruschetta, R. Carli, A. Cenedese.
Motor-level Nonlinear Model Predictive Control for a Tilting Quadrotor. IEEE Conference on Control Technology and Applications (CCTA 2023), pp. 281--286, 2023
Abstract:
This work presents a novel motor-level Nonlinear
Model Predictive Control trajectory tracking controller for an
over-actuated quadrotor with tilting propellers. The proposed
controller directly provides the motor-level commands for both
the tilting and the spinning of the propellers. Moreover, it
optimally solves the control allocation problem arising from
the system’s over-actuation taking into account the physical
constraints of the platform. Leveraging a look-ahead strategy
combined with the knowledge of the actuation limits, the
proposed solution fully exploits the vehicle capabilities and
accurately tracks the desired reference. Simulation results
show that the solution proposed outperforms a state-of-the-art
controller based on Feedback Linearization, in terms of both
trajectory tracking and robustness to unmodeled dynamics.
[
url] [
BibTeX]
N. Lissandrini, L. Battistella, M. Ryll, G. Michieletto, A. Cenedese.
NAPVIG: Local Generalized Voronoi Approximation for Reactive Navigation in Unknown and Dynamic Environments. American Control Conference (ACC), pp. 28-33, 2023
Abstract:
In this paper, we propose a novel online approach
for reactive local navigation of a robotic agent, based on a
fast approximation of the Generalized Voronoi Diagram in a
neighborhood of the robot’s position. We consider the context
of an unknown environment characterized by some narrow pas-
sages and a dynamic configuration. Given the uncertainty and
unpredictability that affect the scenario, we aim at computing
trajectories that are farthest away from every obstacle: this is
obtained by following the Voronoi diagram. To ensure full au-
tonomy, the navigation task is performed relying only upon on-
board sensor measurement without any a-priori knowledge of
the environment. The proposed technique builds upon a smooth
free space representation that is spatially continuous and based
on some raw measurements. In this way, we ensure an efficient
computation of a trajectory that is continuously re-planned
according to incoming sensor data. A theoretical proof shows
that in ideal conditions the outlined solution exactly computes
the local Generalized Voronoi Diagram. Finally, we assess the
reactiveness and precision of the proposed method with realistic
real-time simulations and with real-world experiments.
[
url] [
BibTeX]
C.B. Bran, P. Iob, A. Cenedese, M. Schiavo.
Non-Terminal Sliding Mode Control for a Three-Link Manipulator with Variable Mass. IEEE Conference on Control Technology and Applications (CCTA 2023), pp. 333--338, 2023
Abstract:
Robotic manipulators represent one of the most
useful tools to address repetitive tasks in the industrial en-
vironment. In the primary aluminum industry, for example,
manipulators mounted on vehicles and directly controlled by
human operators are routinely used to feed the potcells with
fluoride and improve the metal generation process. In order
to automatize these operations and limit human presence in
hazardous environments, manipulators can be equipped with
automatic motion control algorithms to perform the requested
tasks. In this work, a solution to this problem is proposed, based
on a non-singular terminal sliding mode control approach and
compared with a classical sliding mode control algorithm. The
devised solution turns out to be efficient and robust, and in the
specific case, able to take into account a non-measurable mass
variation in the manipulator links itself. Numerical simulations
with a multibody tool are employed as a first assessment and
validation of the designed controller.
[
url] [
BibTeX]
A. Fabris, G. Silvello, G.A. Susto, A. Biega.
Pairwise Fairness in Ranking as a Dissatisfaction Measure. ACM International Conference on Web Search and Data Mining, 2023 [
BibTeX]
L. Lorenti, D. Dalle Pezze, J. Andreoli, C. Masiero, N. Gentner, Y. Yang, G.A. Susto.
Predictive Maintenance in the Industry: A Comparative Study on Deep Learning-based Remaining Useful Life Estimation. IEEE International Conference on Industrial Informatics (INDIN), 2023 [
BibTeX]
J. Giordano, A. Cenedese.
Quaternion-Based Non-Singular Terminal Sliding Mode Control for a Satellite-Mounted Space Manipulator. IEEE Control Systems Letters, vol. 7, pp. 2659-2664, 2023
Abstract:
In this letter, a robust control solution for a
satellite equipped with a robotic manipulator is presented.
First, the dynamical model of the system is derived based
on quaternions to describe the evolution of the attitude
of the base satellite. Then, a non-singular terminal sliding
mode controller that employs quaternions for attitude con-
trol, is proposed for concurrently handling all the degrees
of freedom of the system. Moreover, an additional adaptive
term is embedded in the controller to estimate the upper
bounds of disturbances and uncertainties. The result is
a resilient solution able to withstand unmodelled dynam-
ics and interactions. Lyapunov theory is used to prove the
stability of the controller and numerical simulations allow
assessing performance and fuel efficiency.
[
url] [
BibTeX]
L. Cristaldi, P. Esmaili, G. Gruosso, A. La Bella, M. Mecella, R. Scattolini, A. Arman, G.A. Susto, L. Tanca.
The MICS Project. A Data Science Pipeline for Industry 4.0 Applications. 2023 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence, and Neural Engineering (IEEE MetroXRAINE 2023), 2023 [
BibTeX]
D. Marcato, D. Bortolato, V. Martinelli, G. Savarese, G.A. Susto.
Time-Series Deep Learning Anomaly Detection for Particle Accelerators. IFAC World Congress, 2023 [
BibTeX]
A. Beghi, N. Dall'Ora, D. Dalle Pezze, F. Fummi, C. Masiero, S. Spellini, G.A. Susto, F. Tosoni.
VIR2EM: VIrtualization and Remotization for Resilient and Efficient Manufacturing. 26th Forum on specification and Design Languages, 2023 [
BibTeX]
