Publications
21.
Carron, Andrea; Saccani, Danilo; Fagiano, Lorenzo; Zeilinger, Melanie N.
Multi-agent Distributed Model Predictive Control with Connectivity Constraint Conference
vol. 56, no. 2, 2023, (All Open Access, Gold Open Access, Green Open Access).
@conference{Carron20233806,
title = {Multi-agent Distributed Model Predictive Control with Connectivity Constraint},
author = {Andrea Carron and Danilo Saccani and Lorenzo Fagiano and Melanie N. Zeilinger},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85183609109&doi=10.1016%2fj.ifacol.2023.10.1310&partnerID=40&md5=10749559dc883b45c215b676f3a27db8},
doi = {10.1016/j.ifacol.2023.10.1310},
year = {2023},
date = {2023-01-01},
journal = {IFAC-PapersOnLine},
volume = {56},
number = {2},
pages = {3806 – 3811},
abstract = {In cooperative multi-agent robotic systems, coordination is necessary in order to complete a given task. Important examples include search and rescue, operations in hazardous environments, and environmental monitoring. Coordination, in turn, requires simultaneous satisfaction of safety critical constraints, in the form of state and input constraints, and a connectivity constraint, in order to ensure that at every time instant there exists a communication path between every pair of agents in the network. In this work, we present a model predictive controller that tackles the problem of performing multi-agent coordination while simultaneously satisfying safety critical and connectivity constraints. The former is formulated in the form of state and input constraints and the latter as a constraint on the second smallest eigenvalue of the associated communication graph Laplacian matrix, also known as Fiedler eigenvalue, which enforces the connectivity of the communication network. We propose a sequential quadratic programming formulation to solve the associated optimization problem that is amenable to distributed optimization, making the proposed solution suitable for control of multi-agent robotics systems relying on local computation. Finally, the effectiveness of the algorithm is highlighted with a numerical simulation. Copyright © 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)},
note = {All Open Access, Gold Open Access, Green Open Access},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
In cooperative multi-agent robotic systems, coordination is necessary in order to complete a given task. Important examples include search and rescue, operations in hazardous environments, and environmental monitoring. Coordination, in turn, requires simultaneous satisfaction of safety critical constraints, in the form of state and input constraints, and a connectivity constraint, in order to ensure that at every time instant there exists a communication path between every pair of agents in the network. In this work, we present a model predictive controller that tackles the problem of performing multi-agent coordination while simultaneously satisfying safety critical and connectivity constraints. The former is formulated in the form of state and input constraints and the latter as a constraint on the second smallest eigenvalue of the associated communication graph Laplacian matrix, also known as Fiedler eigenvalue, which enforces the connectivity of the communication network. We propose a sequential quadratic programming formulation to solve the associated optimization problem that is amenable to distributed optimization, making the proposed solution suitable for control of multi-agent robotics systems relying on local computation. Finally, the effectiveness of the algorithm is highlighted with a numerical simulation. Copyright © 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
22.
Santos, Pedro Henrique Gomes Dos; Ruiz, Fredy; Fagiano, Lorenzo
Trajectory Planning for Tethered Robots in Uncertain Environments Conference
2023.
@conference{DosSantos2023391,
title = {Trajectory Planning for Tethered Robots in Uncertain Environments},
author = {Pedro Henrique Gomes Dos Santos and Fredy Ruiz and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85177430382&doi=10.1109%2fCoDIT58514.2023.10284096&partnerID=40&md5=6bff713a520c1920f7f05ca8633e4853},
doi = {10.1109/CoDIT58514.2023.10284096},
year = {2023},
date = {2023-01-01},
journal = {9th 2023 International Conference on Control, Decision and Information Technologies, CoDIT 2023},
pages = {391 – 396},
abstract = {The present paper considers the problem of planar motion planning for environments with partially unknown obstacles for a robot with tether connection and missions with multiple target positions. The presence of a tether increases the complexity of the already challenging problem of dynamic motion planning by introducing additional feasibility constraints, i.e. not entangling the cable in any of the obstacles within the working space. For a given mission, the developed algorithm finds a closed trajectory such that after its execution the cable can be retrieved without difficulties. It combines the dynamic path-planning capabilities of the RRTX (Rapid-exploring Random Tree X) method with a geometric approach that efficiently identifies feasible paths that satisfy all the criteria imposed by the problem. Extensive simulation tests demonstrate the validity and reduced computational complexity of the proposed solution. © 2023 IEEE.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
The present paper considers the problem of planar motion planning for environments with partially unknown obstacles for a robot with tether connection and missions with multiple target positions. The presence of a tether increases the complexity of the already challenging problem of dynamic motion planning by introducing additional feasibility constraints, i.e. not entangling the cable in any of the obstacles within the working space. For a given mission, the developed algorithm finds a closed trajectory such that after its execution the cable can be retrieved without difficulties. It combines the dynamic path-planning capabilities of the RRTX (Rapid-exploring Random Tree X) method with a geometric approach that efficiently identifies feasible paths that satisfy all the criteria imposed by the problem. Extensive simulation tests demonstrate the validity and reduced computational complexity of the proposed solution. © 2023 IEEE.
23.
Boffadossi, Roberto; Leonesio, Marco; Fagiano, Lorenzo; Bianchi, Giacomo
Prediction of power consumption from real process data of an industrial wood chip refining plant Conference
vol. 56, no. 2, 2023, (All Open Access, Gold Open Access, Green Open Access).
@conference{Boffadossi20238574,
title = {Prediction of power consumption from real process data of an industrial wood chip refining plant},
author = {Roberto Boffadossi and Marco Leonesio and Lorenzo Fagiano and Giacomo Bianchi},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184963306&doi=10.1016%2fj.ifacol.2023.10.029&partnerID=40&md5=20d75bf1fe3cc1c9d525f676574fed2e},
doi = {10.1016/j.ifacol.2023.10.029},
year = {2023},
date = {2023-01-01},
journal = {IFAC-PapersOnLine},
volume = {56},
number = {2},
pages = {8574 – 8579},
abstract = {Improving the efficiency of production processes is fundamental to minimize their environmental impact and energy consumption. The pulp and paper industry is a highly energy-intensive one that urgently needs to become more efficient, especially in the refining phase. In this framework, the model identification of a wood chips refining process operating in closed loop, pertaining to the production of Medium Density Fiberboard (MDF), is presented here, aimed to provide a long-term prediction of power consumption. We perform the identification via multi-batch Simulation Error Minimization (SEM), employing real process data collected on a large-scale MDF production plant during operation, without using sophisticated models or ad-hoc experimental sessions. The derived model obtains extremely high accuracy on a validation dataset while being simple enough to be used efficiently for production planning optimization. Moreover, it allows us to derive further models to predict the wear of the refiner disc, to be accounted for in a plant optimization procedure as well. Copyright © 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)},
note = {All Open Access, Gold Open Access, Green Open Access},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Improving the efficiency of production processes is fundamental to minimize their environmental impact and energy consumption. The pulp and paper industry is a highly energy-intensive one that urgently needs to become more efficient, especially in the refining phase. In this framework, the model identification of a wood chips refining process operating in closed loop, pertaining to the production of Medium Density Fiberboard (MDF), is presented here, aimed to provide a long-term prediction of power consumption. We perform the identification via multi-batch Simulation Error Minimization (SEM), employing real process data collected on a large-scale MDF production plant during operation, without using sophisticated models or ad-hoc experimental sessions. The derived model obtains extremely high accuracy on a validation dataset while being simple enough to be used efficiently for production planning optimization. Moreover, it allows us to derive further models to predict the wear of the refiner disc, to be accounted for in a plant optimization procedure as well. Copyright © 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
24.
Løwenstein, Kristoffer Fink; Bernardini, Daniele; Fagiano, Lorenzo; Bemporad, Alberto
Physics-informed online learning of gray-box models by moving horizon estimation Journal Article
In: European Journal of Control, vol. 74, 2023, (All Open Access, Green Open Access, Hybrid Gold Open Access).
@article{Løwenstein2023,
title = {Physics-informed online learning of gray-box models by moving horizon estimation},
author = {Kristoffer Fink Løwenstein and Daniele Bernardini and Lorenzo Fagiano and Alberto Bemporad},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85166946697&doi=10.1016%2fj.ejcon.2023.100861&partnerID=40&md5=a00ec01c4e8cc29b88532faefc5ee06d},
doi = {10.1016/j.ejcon.2023.100861},
year = {2023},
date = {2023-01-01},
journal = {European Journal of Control},
volume = {74},
abstract = {A simple yet expressive prediction model is an essential ingredient in model-based control and estimation. Models derived from fundamental physical principles may fail to capture the complexity of the actual system dynamics. A potential solution is the use of a physics-informed, or gray-box model that extends a physics-based model with a data-driven part. Learning the latter might be challenging, due to noisy measurements and lack of full state information. This work presents a method based on Moving Horizon Estimation (MHE) for simultaneous state estimation and training of a black-box submodel, such as a neural network. The method can be used in offline training or applied online for adaptation without any prior knowledge than the white-box submodel. We analyze the capabilities of the method in a two degree of freedom robotic manipulator case study, also showing how it can be used for online adaptation to cope with a time-varying model mismatch. © 2023 The Authors},
note = {All Open Access, Green Open Access, Hybrid Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
A simple yet expressive prediction model is an essential ingredient in model-based control and estimation. Models derived from fundamental physical principles may fail to capture the complexity of the actual system dynamics. A potential solution is the use of a physics-informed, or gray-box model that extends a physics-based model with a data-driven part. Learning the latter might be challenging, due to noisy measurements and lack of full state information. This work presents a method based on Moving Horizon Estimation (MHE) for simultaneous state estimation and training of a black-box submodel, such as a neural network. The method can be used in offline training or applied online for adaptation without any prior knowledge than the white-box submodel. We analyze the capabilities of the method in a two degree of freedom robotic manipulator case study, also showing how it can be used for online adaptation to cope with a time-varying model mismatch. © 2023 The Authors
25.
Lauricella, Marco; Fagiano, Lorenzo
Day-Ahead and Intra-Day Building Load Forecast With Uncertainty Bounds Using Small Data Batches Journal Article
In: IEEE Transactions on Control Systems Technology, vol. 31, no. 6, pp. 2584 – 2595, 2023, (All Open Access, Green Open Access, Hybrid Gold Open Access).
@article{Lauricella20232584,
title = {Day-Ahead and Intra-Day Building Load Forecast With Uncertainty Bounds Using Small Data Batches},
author = {Marco Lauricella and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85161058018&doi=10.1109%2fTCST.2023.3274955&partnerID=40&md5=40c962b12bda3d1d54cedf0318a1a79c},
doi = {10.1109/TCST.2023.3274955},
year = {2023},
date = {2023-01-01},
journal = {IEEE Transactions on Control Systems Technology},
volume = {31},
number = {6},
pages = {2584 – 2595},
abstract = {An approach to provide day-ahead and intra-day load forecasts of buildings, such as electrical or thermal power consumption, is presented. The method aims to obtain a nominal forecast and associated error bounds with small data batches of two weeks for the training phase, resulting in a ready-to-go algorithm that can be employed whenever large datasets of months or years are not available or manageable. These cases include new or renovated constructions, buildings that are subject to changes in purpose and occupants' behavior, or applications on local devices with memory limits. The approach relies on a so-called 'fictitious input' signal to capture the prior information on seasonal and periodic trends of load consumption. Then, linear multistep predictors with different horizon lengths are trained periodically with a small batch of the most recent data, and the associated worst case error bounds are derived, using set membership (SM) methods. Finally, the forecast is computed, for each time step, by intersecting the error bounds of the different multistep predictions and taking the central value of the obtained interval. Such a method is applied here for the first time to real-world data of electrical power consumption of a medium-size building and of cooling power consumption of a large complex. In both cases, the obtained results indicate a tightening of the worst case error bounds between 15% and 25% on average with respect to those obtained with a standard linear SM approach. © 1993-2012 IEEE.},
note = {All Open Access, Green Open Access, Hybrid Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
An approach to provide day-ahead and intra-day load forecasts of buildings, such as electrical or thermal power consumption, is presented. The method aims to obtain a nominal forecast and associated error bounds with small data batches of two weeks for the training phase, resulting in a ready-to-go algorithm that can be employed whenever large datasets of months or years are not available or manageable. These cases include new or renovated constructions, buildings that are subject to changes in purpose and occupants' behavior, or applications on local devices with memory limits. The approach relies on a so-called 'fictitious input' signal to capture the prior information on seasonal and periodic trends of load consumption. Then, linear multistep predictors with different horizon lengths are trained periodically with a small batch of the most recent data, and the associated worst case error bounds are derived, using set membership (SM) methods. Finally, the forecast is computed, for each time step, by intersecting the error bounds of the different multistep predictions and taking the central value of the obtained interval. Such a method is applied here for the first time to real-world data of electrical power consumption of a medium-size building and of cooling power consumption of a large complex. In both cases, the obtained results indicate a tightening of the worst case error bounds between 15% and 25% on average with respect to those obtained with a standard linear SM approach. © 1993-2012 IEEE.
26.
Autelitano, Kevin; Bolognini, Michele; Angelis, Enrico De; Fagiano, Lorenzo; Scaioni, Marco
On the use of drones for vision-based digitalization, diagnostics and energy efficiency assessment of buildings Journal Article
In: Science and Technology for the Built Environment, vol. 29, no. 10, pp. 985 – 997, 2023.
@article{Autelitano2023985,
title = {On the use of drones for vision-based digitalization, diagnostics and energy efficiency assessment of buildings},
author = {Kevin Autelitano and Michele Bolognini and Enrico De Angelis and Lorenzo Fagiano and Marco Scaioni},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85174320853&doi=10.1080%2f23744731.2023.2261807&partnerID=40&md5=f279106b6264aeb6bf5803224042e501},
doi = {10.1080/23744731.2023.2261807},
year = {2023},
date = {2023-01-01},
journal = {Science and Technology for the Built Environment},
volume = {29},
number = {10},
pages = {985 – 997},
abstract = {The adoption of Unmanned Aerial Vehicles for energy efficiency assessment is a promising technique. Although there are some interesting ideas to automate the process, most of the operations are still done manually. An integrated methodology is presented to identify maintenance needs from both RGB and infrared images collected with a commercial drone. The real building selected as case study is reconstructed in a 3D environment through Structure-from-Motion Photogrammetry, while the infrared information is integrated and properly scaled on it to obtain a 3D point cloud model that provides, for each point: (i) information about its position in the space, and (ii) external surface temperature measured during data gathering phase. The point cloud is then segmented into the different sides of the envelope to identify each part of the façade and compared with a model that reconstructs the expected conditions of the building. This procedure enables the development of an automated pipeline to identify defects and failures in a building envelope and to suggest corrective actions. © Copyright © 2023 ASHRAE.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The adoption of Unmanned Aerial Vehicles for energy efficiency assessment is a promising technique. Although there are some interesting ideas to automate the process, most of the operations are still done manually. An integrated methodology is presented to identify maintenance needs from both RGB and infrared images collected with a commercial drone. The real building selected as case study is reconstructed in a 3D environment through Structure-from-Motion Photogrammetry, while the infrared information is integrated and properly scaled on it to obtain a 3D point cloud model that provides, for each point: (i) information about its position in the space, and (ii) external surface temperature measured during data gathering phase. The point cloud is then segmented into the different sides of the envelope to identify each part of the façade and compared with a model that reconstructs the expected conditions of the building. This procedure enables the development of an automated pipeline to identify defects and failures in a building envelope and to suggest corrective actions. © Copyright © 2023 ASHRAE.
27.
Kessler, Nicolas; Fagiano, Lorenzo
On the stabilization of forking and cyclic trajectories for nonlinear systems Conference
vol. 56, no. 3, 2023, (All Open Access, Gold Open Access).
@conference{Kessler2023199,
title = {On the stabilization of forking and cyclic trajectories for nonlinear systems},
author = {Nicolas Kessler and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184348541&doi=10.1016%2fj.ifacol.2023.12.024&partnerID=40&md5=b8d1f378c3f815a62cc0667f5df9af21},
doi = {10.1016/j.ifacol.2023.12.024},
year = {2023},
date = {2023-01-01},
journal = {IFAC-PapersOnLine},
volume = {56},
number = {3},
pages = {199 – 204},
abstract = {Stabilizing a reference trajectory for a nonlinear system is a common, non-trivial task in control theory. An approach to solve this problem is to approximate the nonlinear system along the trajectory as an uncertain linear time-varying one, and to solve an optimization problem featuring Linear Matrix Inequality (LMI) constraints to derive a stabilizing, smooth, gain-scheduled control law. Such an approach is extended here by considering a set of reference trajectories instead of a single one, such that switching among them is permitted. These switching events are commonly encountered in industrial plants, such as energy generation systems, and are of high relevance in practice. The approach allows one to derive a gain-scheduled control law guaranteeing asymptotic stability also during the switching and accounting for the linearization errors. Simulation results on a chemical system highlight the effectiveness of the method. Copyright © 2023 The Authors.},
note = {All Open Access, Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Stabilizing a reference trajectory for a nonlinear system is a common, non-trivial task in control theory. An approach to solve this problem is to approximate the nonlinear system along the trajectory as an uncertain linear time-varying one, and to solve an optimization problem featuring Linear Matrix Inequality (LMI) constraints to derive a stabilizing, smooth, gain-scheduled control law. Such an approach is extended here by considering a set of reference trajectories instead of a single one, such that switching among them is permitted. These switching events are commonly encountered in industrial plants, such as energy generation systems, and are of high relevance in practice. The approach allows one to derive a gain-scheduled control law guaranteeing asymptotic stability also during the switching and accounting for the linearization errors. Simulation results on a chemical system highlight the effectiveness of the method. Copyright © 2023 The Authors.
28.
Sabug, Lorenzo; Ruiz, Fredy; Fagiano, Lorenzo
A Set Membership approach to black-box optimization for time-varying problems Conference
vol. 56, no. 2, 2023, (All Open Access, Gold Open Access, Green Open Access).
@conference{Sabug20233966,
title = {A Set Membership approach to black-box optimization for time-varying problems},
author = {Lorenzo Sabug and Fredy Ruiz and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85184961435&doi=10.1016%2fj.ifacol.2023.10.1343&partnerID=40&md5=adb28890b74d1827ad7e6a67a617beaa},
doi = {10.1016/j.ifacol.2023.10.1343},
year = {2023},
date = {2023-01-01},
journal = {IFAC-PapersOnLine},
volume = {56},
number = {2},
pages = {3966 – 3971},
abstract = {A novel method to tackle black-box optimization for time-varying problems is proposed. Using a Set Membership (SM) framework, the approach directly adjusts the uncertainty associated with old data points as new samples are introduced. Uninformative old samples are discarded, and the adjusted model guides the exploitation and exploration routines as characteristic of black-box optimization. With the proposed method, there is no need to estimate the time-related rate of change of the hidden function, as required in previous literature. We provide results of a benchmark test, comparing the performance of the proposed method to other approaches to time-varying black-box optimization, with promising results. Copyright © 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)},
note = {All Open Access, Gold Open Access, Green Open Access},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
A novel method to tackle black-box optimization for time-varying problems is proposed. Using a Set Membership (SM) framework, the approach directly adjusts the uncertainty associated with old data points as new samples are introduced. Uninformative old samples are discarded, and the adjusted model guides the exploitation and exploration routines as characteristic of black-box optimization. With the proposed method, there is no need to estimate the time-related rate of change of the hidden function, as required in previous literature. We provide results of a benchmark test, comparing the performance of the proposed method to other approaches to time-varying black-box optimization, with promising results. Copyright © 2023 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)
29.
Sabug, Lorenzo; Incremona, Gian Paolo; Tanelli, Mara; Ruiz, Fredy; Fagiano, Lorenzo
Simultaneous design of passive and active spacecraft attitude control using black-box optimization Journal Article
In: Control Engineering Practice, vol. 135, 2023, (All Open Access, Hybrid Gold Open Access).
@article{Sabug2023,
title = {Simultaneous design of passive and active spacecraft attitude control using black-box optimization},
author = {Lorenzo Sabug and Gian Paolo Incremona and Mara Tanelli and Fredy Ruiz and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85151331587&doi=10.1016%2fj.conengprac.2023.105516&partnerID=40&md5=895737082534765027bbe8d0fc60b771},
doi = {10.1016/j.conengprac.2023.105516},
year = {2023},
date = {2023-01-01},
journal = {Control Engineering Practice},
volume = {135},
abstract = {This paper investigates the simultaneous design of active attitude control and passive attitude compensation mechanism for a spacecraft to satisfy practically-motivated mission objectives and constraints. The expressions of these fitness-related metrics with respect to the design variables are not analytically available, due to the nontrivial interactions between the spacecraft components and the interactions with the environment. Thus, such functions can only be approximately learned from data derived from simulations. We approach this difficult design problem using a black-box optimization (BBO)-based approach, which combines learning and optimizing the objective and constraint functions by design of experiments. The proposed BBO-based approach is assessed in the context of a 3U CubeSat system design with both a passive magnetic attitude compensation and an active reaction wheel-based control, tested on a simulator considering orbital and environmental dynamics. Simulation results and statistical tests compared to other design methods show the capability of the BBO-based approach to provide a design with the best tracking performance while at the same time satisfying ground station communication requirements and power budget. © 2023 Elsevier Ltd},
note = {All Open Access, Hybrid Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This paper investigates the simultaneous design of active attitude control and passive attitude compensation mechanism for a spacecraft to satisfy practically-motivated mission objectives and constraints. The expressions of these fitness-related metrics with respect to the design variables are not analytically available, due to the nontrivial interactions between the spacecraft components and the interactions with the environment. Thus, such functions can only be approximately learned from data derived from simulations. We approach this difficult design problem using a black-box optimization (BBO)-based approach, which combines learning and optimizing the objective and constraint functions by design of experiments. The proposed BBO-based approach is assessed in the context of a 3U CubeSat system design with both a passive magnetic attitude compensation and an active reaction wheel-based control, tested on a simulator considering orbital and environmental dynamics. Simulation results and statistical tests compared to other design methods show the capability of the BBO-based approach to provide a design with the best tracking performance while at the same time satisfying ground station communication requirements and power budget. © 2023 Elsevier Ltd
30.
Saccani, Danilo; Cecchin, Leonardo; Fagiano, Lorenzo
Multitrajectory Model Predictive Control for Safe UAV Navigation in an Unknown Environment Journal Article
In: IEEE Transactions on Control Systems Technology, vol. 31, no. 5, pp. 1982 – 1997, 2023, (All Open Access, Green Open Access, Hybrid Gold Open Access).
@article{Saccani20231982,
title = {Multitrajectory Model Predictive Control for Safe UAV Navigation in an Unknown Environment},
author = {Danilo Saccani and Leonardo Cecchin and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85141627994&doi=10.1109%2fTCST.2022.3216989&partnerID=40&md5=ca5d3d5501ff1efd806aa7c9932fc915},
doi = {10.1109/TCST.2022.3216989},
year = {2023},
date = {2023-01-01},
journal = {IEEE Transactions on Control Systems Technology},
volume = {31},
number = {5},
pages = {1982 – 1997},
abstract = {The problem of navigating an unmanned aerial vehicle (UAV) in an unknown environment is addressed with a novel model predictive control (MPC) formulation, named multitrajectory MPC (mt-MPC). The objective is to safely drive the vehicle to the desired target location by relying only on the partial description of the surroundings provided by an exteroceptive sensor. This information results in time-varying constraints during the navigation among obstacles. The proposed mt-MPC generates a sequence of position set points that are fed to control loops at lower hierarchical levels. To do so, the mt-MPC predicts two different state trajectories, a safe one and an exploiting one, in the same finite horizon optimal control problem (FHOCP). This formulation, particularly suitable for problems with uncertain time-varying constraints, allows one to partially decouple constraint satisfaction (safety) from cost function minimization (exploitation). Uncertainty due to modeling errors and sensors noise is taken into account as well, in a set membership (SM) framework. Theoretical guarantees of persistent obstacle avoidance are derived under suitable assumptions, and the approach is demonstrated experimentally out-of-the-laboratory on a prototype built with off-the-shelf components. © 1993-2012 IEEE.},
note = {All Open Access, Green Open Access, Hybrid Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
The problem of navigating an unmanned aerial vehicle (UAV) in an unknown environment is addressed with a novel model predictive control (MPC) formulation, named multitrajectory MPC (mt-MPC). The objective is to safely drive the vehicle to the desired target location by relying only on the partial description of the surroundings provided by an exteroceptive sensor. This information results in time-varying constraints during the navigation among obstacles. The proposed mt-MPC generates a sequence of position set points that are fed to control loops at lower hierarchical levels. To do so, the mt-MPC predicts two different state trajectories, a safe one and an exploiting one, in the same finite horizon optimal control problem (FHOCP). This formulation, particularly suitable for problems with uncertain time-varying constraints, allows one to partially decouple constraint satisfaction (safety) from cost function minimization (exploitation). Uncertainty due to modeling errors and sensors noise is taken into account as well, in a set membership (SM) framework. Theoretical guarantees of persistent obstacle avoidance are derived under suitable assumptions, and the approach is demonstrated experimentally out-of-the-laboratory on a prototype built with off-the-shelf components. © 1993-2012 IEEE.
31.
Bolognini, Michele; Fagiano, Lorenzo; Limongelli, Maria Pina
A fault-tolerant automatic mission planner for a fleet of aerial vehicles Journal Article
In: Control Engineering Practice, vol. 135, 2023.
@article{Bolognini2023,
title = {A fault-tolerant automatic mission planner for a fleet of aerial vehicles},
author = {Michele Bolognini and Lorenzo Fagiano and Maria Pina Limongelli},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85151426397&doi=10.1016%2fj.conengprac.2023.105501&partnerID=40&md5=15adfaae41c674fb698b1a46919b5eab},
doi = {10.1016/j.conengprac.2023.105501},
year = {2023},
date = {2023-01-01},
journal = {Control Engineering Practice},
volume = {135},
abstract = {Unmanned Aerial Vehicles are versatile tools for inspection tasks, for example of the built environment. Due to their limited flight time, it is sensible to employ multiple units to decrease the total mission time and avoid additional trips to change batteries. The use of more units makes the mission planning more complex; moreover, a fault tolerant plan that is resilient at least to a single failure is desirable. To solve these problems, a new algorithm is proposed, which, via hierarchical decomposition and numerical optimization, effectively deals with: (1) the efficient generation of a suitable path for each drone, and (2) guaranteeing mission robustness against a single fault. The automated generation and clustering of points of interest to be visited is addressed, too, as part of the whole procedure. Using accurate models of two real buildings, it is shown that the approach delivers close-to-optimal solutions with small computational time, thus being compatible with real-world operation. © 2023 Elsevier Ltd},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Unmanned Aerial Vehicles are versatile tools for inspection tasks, for example of the built environment. Due to their limited flight time, it is sensible to employ multiple units to decrease the total mission time and avoid additional trips to change batteries. The use of more units makes the mission planning more complex; moreover, a fault tolerant plan that is resilient at least to a single failure is desirable. To solve these problems, a new algorithm is proposed, which, via hierarchical decomposition and numerical optimization, effectively deals with: (1) the efficient generation of a suitable path for each drone, and (2) guaranteeing mission robustness against a single fault. The automated generation and clustering of points of interest to be visited is addressed, too, as part of the whole procedure. Using accurate models of two real buildings, it is shown that the approach delivers close-to-optimal solutions with small computational time, thus being compatible with real-world operation. © 2023 Elsevier Ltd
32.
Boffadossi, Roberto; Bonassi, Fabio; Fagiano, Lorenzo; Scattolini, Riccardo; Cataldo, Andrea
Safeguarded optimal policy learning for a smart discrete manufacturing plant Conference
vol. 55, no. 2, 2022, (All Open Access, Gold Open Access).
@conference{Boffadossi2022396,
title = {Safeguarded optimal policy learning for a smart discrete manufacturing plant},
author = {Roberto Boffadossi and Fabio Bonassi and Lorenzo Fagiano and Riccardo Scattolini and Andrea Cataldo},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85132176041&doi=10.1016%2fj.ifacol.2022.04.226&partnerID=40&md5=bc9a1b128effb736ea5685310325bf97},
doi = {10.1016/j.ifacol.2022.04.226},
year = {2022},
date = {2022-01-01},
journal = {IFAC-PapersOnLine},
volume = {55},
number = {2},
pages = {396 – 401},
abstract = {An approach to safely learn and deploy, at fast rate, a given optimization-based controller for the routing problem in smart manufacturing is presented. The considered application features a large number of integer decision variables, combined with nonlinear dynamics, temporal-logic constraints, and hard safety constraints. The approach employs a neural network as feedback controller, trained using a data-set of state-input pairs collected with the optimization-based controller. A safeguard mechanism checks whether the input computed by the neural network is feasible or not, in the latter case the optimization-based controller is called. Results on a high-fidelity simulation suite indicate a strong decrease of average computational time combined with a negligible loss of plant performance. © 2022 Elsevier B.V.. All rights reserved.},
note = {All Open Access, Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
An approach to safely learn and deploy, at fast rate, a given optimization-based controller for the routing problem in smart manufacturing is presented. The considered application features a large number of integer decision variables, combined with nonlinear dynamics, temporal-logic constraints, and hard safety constraints. The approach employs a neural network as feedback controller, trained using a data-set of state-input pairs collected with the optimization-based controller. A safeguard mechanism checks whether the input computed by the neural network is feasible or not, in the latter case the optimization-based controller is called. Results on a high-fidelity simulation suite indicate a strong decrease of average computational time combined with a negligible loss of plant performance. © 2022 Elsevier B.V.. All rights reserved.
33.
Fagiano, Lorenzo; Quack, Manfred; Bauer, Florian; Carnel, Lode; Oland, Espen
Autonomous Airborne Wind Energy Systems: Accomplishments and Challenges Journal Article
In: Annual Review of Control, Robotics, and Autonomous Systems, vol. 5, pp. 603 – 631, 2022.
@article{Fagiano2022603,
title = {Autonomous Airborne Wind Energy Systems: Accomplishments and Challenges},
author = {Lorenzo Fagiano and Manfred Quack and Florian Bauer and Lode Carnel and Espen Oland},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85129871786&doi=10.1146%2fannurev-control-042820-124658&partnerID=40&md5=4afdc008d81bc65f78fbfceb8a8deb63},
doi = {10.1146/annurev-control-042820-124658},
year = {2022},
date = {2022-01-01},
journal = {Annual Review of Control, Robotics, and Autonomous Systems},
volume = {5},
pages = {603 – 631},
abstract = {Airborne wind energy (AWE) is a fascinating technology to convert wind power into electricity with an autonomous tethered aircraft. Deemed a potentially game-changing solution, AWE is attracting the attention of policy makers and stakeholders with the promise of producing large amounts of cost-competitive electricity with wide applicability worldwide. Since the pioneering experimental endeavors in the years 2000-2010, there has been a clear technology convergence trend and steady progress in the field. Today, AWE systems can operate automatically with minimal supervision in all operational phases. A first product is also being commercialized. However, all-Around fully autonomous operation still presents important fundamental challenges that are conceptually similar to those of other systems that promise to change our lives, such as fully autonomous passenger cars or service drones. At the same time, autonomous operation is necessary to enable large-scale AWE, thus combining challenging fundamental problems with high potential impact on society and the economy. This article describes the state of the art of this technology from a system perspective and witha critical view on some fundamental aspects, presents the latest automatic control results by prominent industrial players, and finally points out the most important challenges on the road to fully autonomous AWE systems. Copyright © 2022 by Annual Reviews.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Airborne wind energy (AWE) is a fascinating technology to convert wind power into electricity with an autonomous tethered aircraft. Deemed a potentially game-changing solution, AWE is attracting the attention of policy makers and stakeholders with the promise of producing large amounts of cost-competitive electricity with wide applicability worldwide. Since the pioneering experimental endeavors in the years 2000-2010, there has been a clear technology convergence trend and steady progress in the field. Today, AWE systems can operate automatically with minimal supervision in all operational phases. A first product is also being commercialized. However, all-Around fully autonomous operation still presents important fundamental challenges that are conceptually similar to those of other systems that promise to change our lives, such as fully autonomous passenger cars or service drones. At the same time, autonomous operation is necessary to enable large-scale AWE, thus combining challenging fundamental problems with high potential impact on society and the economy. This article describes the state of the art of this technology from a system perspective and witha critical view on some fundamental aspects, presents the latest automatic control results by prominent industrial players, and finally points out the most important challenges on the road to fully autonomous AWE systems. Copyright © 2022 by Annual Reviews.
34.
Paganelli, Sofia; Yunus, Ilhan; Fagiano, Lorenzo
2022.
@conference{Paganelli2022664,
title = {Comfort-Aware Trajectory Planning in Autonomous Driving via Multi-Objective Nonlinear Model Predictive Control},
author = {Sofia Paganelli and Ilhan Yunus and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85144598135&doi=10.1109%2fCCTA49430.2022.9966123&partnerID=40&md5=295bc9aa3daf718ea14d6e62632415ac},
doi = {10.1109/CCTA49430.2022.9966123},
year = {2022},
date = {2022-01-01},
journal = {2022 IEEE Conference on Control Technology and Applications, CCTA 2022},
pages = {664 – 669},
abstract = {Autonomous passenger vehicles are expected to change mobility as we know it. They can potentially offer many advantages over traditional vehicles, but user acceptance is fundamental to benefit from them. In particular, comfort will be one of the most relevant deciding factors. The lack of control over the vehicle's motion will increase the passengers' inability to predict the path, raising their discomfort. The most straightforward way to maximize comfort is to adopt smooth and slow speed and cornering profiles, but the risk is to penalize too much the total travel time. A preliminary study on the optimal trajectory planning for an autonomous vehicle to find the best user-defined trade-off between comfort and travel time is presented. The problem is approached with a Nonlinear Model-Predictive Control strategy. After describing a suitable vehicle model, the complete formulation of the algorithm is presented. A multi-objective analysis is then carried out to obtain the Pareto Front and the Utopian solution for comfort and travel time. Finally, the algorithm is applied to a case study, assessing the overall performance. © 2022 IEEE.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Autonomous passenger vehicles are expected to change mobility as we know it. They can potentially offer many advantages over traditional vehicles, but user acceptance is fundamental to benefit from them. In particular, comfort will be one of the most relevant deciding factors. The lack of control over the vehicle's motion will increase the passengers' inability to predict the path, raising their discomfort. The most straightforward way to maximize comfort is to adopt smooth and slow speed and cornering profiles, but the risk is to penalize too much the total travel time. A preliminary study on the optimal trajectory planning for an autonomous vehicle to find the best user-defined trade-off between comfort and travel time is presented. The problem is approached with a Nonlinear Model-Predictive Control strategy. After describing a suitable vehicle model, the complete formulation of the algorithm is presented. A multi-objective analysis is then carried out to obtain the Pareto Front and the Utopian solution for comfort and travel time. Finally, the algorithm is applied to a case study, assessing the overall performance. © 2022 IEEE.
35.
Mohammed, Tareg; Fagiano, Lorenzo
Fault- Tolerant Control of a Tethered Aircraft for Airborne Wind Energy Conference
2022.
@conference{Mohammed2022279,
title = {Fault- Tolerant Control of a Tethered Aircraft for Airborne Wind Energy},
author = {Tareg Mohammed and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85144595654&doi=10.1109%2fCCTA49430.2022.9966057&partnerID=40&md5=ff6d6ca86b40c956ecfb45fc08cb73c2},
doi = {10.1109/CCTA49430.2022.9966057},
year = {2022},
date = {2022-01-01},
journal = {2022 IEEE Conference on Control Technology and Applications, CCTA 2022},
pages = {279 – 284},
abstract = {Reliability and fault tolerance are crucial aspects for the industrialization of Airborne Wind Energy (AWE) systems, yet the scientific literature on this topic is still scarce. A study on fault tolerant control of a hybrid rigid-wing/quad- copter tethered aircraft used in AWE is presented. The fault tolerant control structure features a combination of passive measures. The former combine daisy chain control allocation with Internal Model Control (IMC), enabling the system to immediately counteract saturations or faults of one or more actuators. The active measure employs a quantitative model-based fault detection and isolation (FDI) approach to identify a failure in one or more of the discrete control surfaces (rudder, ailerons, elevator). The faulty actuator is then excluded from the control allocation strategy. The FDI approach is based on a residual indicator of the discrepancy between the actual system behavior and the one predicted by a dynamical model fed by the commanded control signals. Simulations of a realistic model of the tethered aircraft show promising performance of the method. © 2022 IEEE.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Reliability and fault tolerance are crucial aspects for the industrialization of Airborne Wind Energy (AWE) systems, yet the scientific literature on this topic is still scarce. A study on fault tolerant control of a hybrid rigid-wing/quad- copter tethered aircraft used in AWE is presented. The fault tolerant control structure features a combination of passive measures. The former combine daisy chain control allocation with Internal Model Control (IMC), enabling the system to immediately counteract saturations or faults of one or more actuators. The active measure employs a quantitative model-based fault detection and isolation (FDI) approach to identify a failure in one or more of the discrete control surfaces (rudder, ailerons, elevator). The faulty actuator is then excluded from the control allocation strategy. The FDI approach is based on a residual indicator of the discrepancy between the actual system behavior and the one predicted by a dynamical model fed by the commanded control signals. Simulations of a realistic model of the tethered aircraft show promising performance of the method. © 2022 IEEE.
36.
Terzi, Enrico; Farina, Marcello; Fagiano, Lorenzo; Scattolini, Riccardo
Robust multi-rate predictive control using multi-step prediction models learned from data Journal Article
In: Automatica, vol. 136, 2022, (All Open Access, Green Open Access).
@article{Terzi2022,
title = {Robust multi-rate predictive control using multi-step prediction models learned from data},
author = {Enrico Terzi and Marcello Farina and Lorenzo Fagiano and Riccardo Scattolini},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85114444637&doi=10.1016%2fj.automatica.2021.109852&partnerID=40&md5=3d0234d28014a15d091149596ea7221e},
doi = {10.1016/j.automatica.2021.109852},
year = {2022},
date = {2022-01-01},
journal = {Automatica},
volume = {136},
abstract = {This note extends a recently proposed algorithm for model identification and robust model predictive control (MPC) of asymptotically stable, linear time-invariant systems subject to process and measurement disturbances. Independent output predictors for different horizon values are estimated with Set Membership methods. It is shown that the corresponding prediction error bounds are the least conservative in the considered model class. Then, a new multi-rate robust MPC algorithm is developed, employing said multi-step predictors to robustly enforce constraints and stability against disturbances and model uncertainty, and to reduce conservativeness. A simulation example illustrates the effectiveness of the approach. © 2021 Elsevier Ltd},
note = {All Open Access, Green Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
This note extends a recently proposed algorithm for model identification and robust model predictive control (MPC) of asymptotically stable, linear time-invariant systems subject to process and measurement disturbances. Independent output predictors for different horizon values are estimated with Set Membership methods. It is shown that the corresponding prediction error bounds are the least conservative in the considered model class. Then, a new multi-rate robust MPC algorithm is developed, employing said multi-step predictors to robustly enforce constraints and stability against disturbances and model uncertainty, and to reduce conservativeness. A simulation example illustrates the effectiveness of the approach. © 2021 Elsevier Ltd
37.
Galbiati, Raffaele; Sabug, Lorenzo; Ruiz, Fredy; Fagiano, Lorenzo
Direct control design using a Set Membership-based black-box optimization approach Conference
2022.
@conference{Galbiati20221259,
title = {Direct control design using a Set Membership-based black-box optimization approach},
author = {Raffaele Galbiati and Lorenzo Sabug and Fredy Ruiz and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85144593184&doi=10.1109%2fCCTA49430.2022.9966147&partnerID=40&md5=4c759264fead5c01a6084c9ba8322b43},
doi = {10.1109/CCTA49430.2022.9966147},
year = {2022},
date = {2022-01-01},
journal = {2022 IEEE Conference on Control Technology and Applications, CCTA 2022},
pages = {1259 – 1264},
abstract = {The problem of controller tuning is a challenging question for practitioners, especially when industrial plants are involved. In such plants, different physical mechanisms come into play and may interact with one another in a complex manner. This aspect, together with uncertainty due to measurement errors and the environment, makes it difficult to derive an accurate plant model, and even more to optimally tune a controller without extensive trial-and-error procedures. To alleviate this problem, an automated control tuning setup is proposed, built on a recently proposed black-box optimization method. The approach iteratively chooses the next controller parameter values, given the data from previous experiments. The proposed approach is tested on a blower-based disk levitation system, after implementing several solutions to enable its real-time usage. In particular, a supervisory logic with an experiment stopping mechanism and a fallback solution is adopted, to save time on experiments with parameters that lead to close-loop instability. The effectiveness of the proposed automated tuning setup is demonstrated, and the results are compared with those of a controller designed from an estimated plant model. © 2022 IEEE.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
The problem of controller tuning is a challenging question for practitioners, especially when industrial plants are involved. In such plants, different physical mechanisms come into play and may interact with one another in a complex manner. This aspect, together with uncertainty due to measurement errors and the environment, makes it difficult to derive an accurate plant model, and even more to optimally tune a controller without extensive trial-and-error procedures. To alleviate this problem, an automated control tuning setup is proposed, built on a recently proposed black-box optimization method. The approach iteratively chooses the next controller parameter values, given the data from previous experiments. The proposed approach is tested on a blower-based disk levitation system, after implementing several solutions to enable its real-time usage. In particular, a supervisory logic with an experiment stopping mechanism and a fallback solution is adopted, to save time on experiments with parameters that lead to close-loop instability. The effectiveness of the proposed automated tuning setup is demonstrated, and the results are compared with those of a controller designed from an estimated plant model. © 2022 IEEE.
38.
Leonesio, Marco; Fagiano, Lorenzo
A semi-supervised physics-informed classifier for centerless grinding operations Conference
2022.
@conference{Leonesio2022977,
title = {A semi-supervised physics-informed classifier for centerless grinding operations},
author = {Marco Leonesio and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85144590634&doi=10.1109%2fCCTA49430.2022.9966132&partnerID=40&md5=e9dd977caa5a52a6b6fdded3d0e2e5d1},
doi = {10.1109/CCTA49430.2022.9966132},
year = {2022},
date = {2022-01-01},
journal = {2022 IEEE Conference on Control Technology and Applications, CCTA 2022},
pages = {977 – 982},
abstract = {Centerless grinding is a machining process characterized by highly nonlinear dynamics and large model uncertainty, making it difficult to predict the quality of the worked parts on the basis of the chosen process parameters. Indeed, it is shown that both physics-based and learning-based approaches alone achieve non-satisfactory prediction performance. In this paper a physics-informed learning approach for this problem is presented. It exploits both the prediction of a physics-based (PB) simulation model and a reduced set of experimental data for a data-driven correction. The approach relies on a hierarchical semi-supervised classification, where the training data, classified on the basis of the three quality intervals of interest, are divided in a certain number of sub-clusters w.r.t. the process input parameters (primary features) and enhanced with the classification prediction provided by a physics-based model (apriori knowledge injection). These sub-clusters are then used in the prediction phase, either directly or through a support vector machine predictor. The results on synthetic data provided by a high-fidelity model show an accuracy (Correct Classification Rate) of 97%, vs 94 % of black-box learning methods and 81% of the physics-based model alone. © 2022 IEEE.},
keywords = {},
pubstate = {published},
tppubtype = {conference}
}
Centerless grinding is a machining process characterized by highly nonlinear dynamics and large model uncertainty, making it difficult to predict the quality of the worked parts on the basis of the chosen process parameters. Indeed, it is shown that both physics-based and learning-based approaches alone achieve non-satisfactory prediction performance. In this paper a physics-informed learning approach for this problem is presented. It exploits both the prediction of a physics-based (PB) simulation model and a reduced set of experimental data for a data-driven correction. The approach relies on a hierarchical semi-supervised classification, where the training data, classified on the basis of the three quality intervals of interest, are divided in a certain number of sub-clusters w.r.t. the process input parameters (primary features) and enhanced with the classification prediction provided by a physics-based model (apriori knowledge injection). These sub-clusters are then used in the prediction phase, either directly or through a support vector machine predictor. The results on synthetic data provided by a high-fidelity model show an accuracy (Correct Classification Rate) of 97%, vs 94 % of black-box learning methods and 81% of the physics-based model alone. © 2022 IEEE.
39.
Bolognini, Michele; Izzo, Giovanni; Marchisotti, Daniele; Fagiano, Lorenzo; Limongelli, Maria Pina; Zappa, Emanuele
Vision-based modal analysis of built environment structures with multiple drones Journal Article
In: Automation in Construction, vol. 143, 2022, (All Open Access, Green Open Access).
@article{Bolognini2022,
title = {Vision-based modal analysis of built environment structures with multiple drones},
author = {Michele Bolognini and Giovanni Izzo and Daniele Marchisotti and Lorenzo Fagiano and Maria Pina Limongelli and Emanuele Zappa},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85137271593&doi=10.1016%2fj.autcon.2022.104550&partnerID=40&md5=6d20f5731bab019c50ae73b2e2e48843},
doi = {10.1016/j.autcon.2022.104550},
year = {2022},
date = {2022-01-01},
journal = {Automation in Construction},
volume = {143},
abstract = {Unmanned Aerial Vehicles are employed for vision-based modal analysis of civil infrastructure, as they overcome range limitations of fixed cameras and measure the oscillations of a structure up close. Nevertheless, their potential is not fully exploited: they are often piloted manually and one at a time, though one drone is unable to capture high resolution displacement of a whole structure. An approach is explored here, employing multiple drones simultaneously to estimate natural frequencies and modal shapes of a structure, by synchronizing their measurement. The ability of the method to detect modal parameter variations is assessed, such that it can identify anomalies in the structure. Procedures are applied to a test structure, yielding maximum natural frequency estimation errors of 0.2% with respect to accelerometers. The results suggest the accuracy of the approach is high enough to warrant further development and support autonomous, multi-drone applications to the inspection of the built environment. © 2022 Elsevier B.V.},
note = {All Open Access, Green Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Unmanned Aerial Vehicles are employed for vision-based modal analysis of civil infrastructure, as they overcome range limitations of fixed cameras and measure the oscillations of a structure up close. Nevertheless, their potential is not fully exploited: they are often piloted manually and one at a time, though one drone is unable to capture high resolution displacement of a whole structure. An approach is explored here, employing multiple drones simultaneously to estimate natural frequencies and modal shapes of a structure, by synchronizing their measurement. The ability of the method to detect modal parameter variations is assessed, such that it can identify anomalies in the structure. Procedures are applied to a test structure, yielding maximum natural frequency estimation errors of 0.2% with respect to accelerometers. The results suggest the accuracy of the approach is high enough to warrant further development and support autonomous, multi-drone applications to the inspection of the built environment. © 2022 Elsevier B.V.
40.
Sabug, Lorenzo; Ruiz, Fredy; Fagiano, Lorenzo
SMGO-Δ: Balancing caution and reward in global optimization with black-box constraints Journal Article
In: Information Sciences, vol. 605, pp. 15 – 42, 2022, (All Open Access, Green Open Access, Hybrid Gold Open Access).
@article{Sabug202215,
title = {SMGO-Δ: Balancing caution and reward in global optimization with black-box constraints},
author = {Lorenzo Sabug and Fredy Ruiz and Lorenzo Fagiano},
url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85130411416&doi=10.1016%2fj.ins.2022.05.017&partnerID=40&md5=4f4ab2b8bc54e9717bf2550c55a99896},
doi = {10.1016/j.ins.2022.05.017},
year = {2022},
date = {2022-01-01},
journal = {Information Sciences},
volume = {605},
pages = {15 – 42},
abstract = {In numerous applications across all science and engineering areas, there are optimization problems where both the objective function and the constraints have no closed-form expression or are too complex to be managed analytically, so that they can only be evaluated through experiments. To address such issues, we design a global optimization technique for problems with black-box objective and constraints. Assuming Lipschitz continuity of the cost and constraint functions, a Set Membership framework is adopted to build a surrogate model of the optimization program, that is used for exploitation and exploration routines. The resulting algorithm, named Set Membership Global Optimization with black-box constraints (SMGO-Δ), features one tunable risk parameter, which the user can intuitively adjust to trade-off safety, exploitation, and exploration. The theoretical properties of the algorithm are derived, and the optimization performance is compared with representative techniques from the literature in several benchmarks. An extension to uncertain cost/constraint function outcomes is presented, too, as well as computational aspects. Lastly, the approach is tested and compared with constrained Bayesian optimization in a case study pertaining to model predictive control tuning for a servomechanism with disturbances and plant uncertainties, addressing practically-motivated task-level constraints. © 2022 Elsevier Inc.},
note = {All Open Access, Green Open Access, Hybrid Gold Open Access},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In numerous applications across all science and engineering areas, there are optimization problems where both the objective function and the constraints have no closed-form expression or are too complex to be managed analytically, so that they can only be evaluated through experiments. To address such issues, we design a global optimization technique for problems with black-box objective and constraints. Assuming Lipschitz continuity of the cost and constraint functions, a Set Membership framework is adopted to build a surrogate model of the optimization program, that is used for exploitation and exploration routines. The resulting algorithm, named Set Membership Global Optimization with black-box constraints (SMGO-Δ), features one tunable risk parameter, which the user can intuitively adjust to trade-off safety, exploitation, and exploration. The theoretical properties of the algorithm are derived, and the optimization performance is compared with representative techniques from the literature in several benchmarks. An extension to uncertain cost/constraint function outcomes is presented, too, as well as computational aspects. Lastly, the approach is tested and compared with constrained Bayesian optimization in a case study pertaining to model predictive control tuning for a servomechanism with disturbances and plant uncertainties, addressing practically-motivated task-level constraints. © 2022 Elsevier Inc.