{"id":159,"date":"2021-07-31T14:49:10","date_gmt":"2021-07-31T14:49:10","guid":{"rendered":"https:\/\/elo-x.eu\/?page_id=159"},"modified":"2021-08-01T14:31:46","modified_gmt":"2021-08-01T14:31:46","slug":"publications","status":"publish","type":"page","link":"https:\/\/elo-x.eu\/?page_id=159","title":{"rendered":"Publications"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Publications<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t\n\t\t\t\t\t\t<\/span>\n\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
<\/a>
<\/div>
\r\n
<\/div><\/div><\/form>
115 entries<\/span> «<\/a> ‹<\/a> 1 of 6 ›<\/a> »<\/a> <\/div><\/div>
1.<\/div>

Gottardini, Andrea; Cecchin, Leonardo; Demir, Ozan; Fagiano, Lorenzo<\/p>

Data-Driven Nonlinear Model Predictive Control for Grading Functions for Excavators Working paper<\/span> Forthcoming<\/span><\/p>

Forthcoming.<\/p>

Abstract<\/a><\/span> | BibTeX<\/a><\/span><\/p>

@workingpaper{gottardini_data-driven_2025,
\r\ntitle = {Data-Driven Nonlinear Model Predictive Control for Grading Functions for Excavators},
\r\nauthor = {Andrea Gottardini and Leonardo Cecchin and Ozan Demir and Lorenzo Fagiano},
\r\nyear  = {2025},
\r\ndate = {2025-12-31},
\r\nabstract = {Hydraulic excavators are essential construc-
\r\ntion machines widely utilized for ground shaping tasks,
\r\nsuch as horizontal leveling and creating sloped surfaces.
\r\nThese operations require a high degree of precision, which
\r\ncan be challenging for unskilled workers.
\r\nThe implementation of automation in hydraulic machin-
\r\nery has the potential to significantly enhance productivity
\r\nby improving accuracy and reducing reliance on highly
\r\ntrained labor. However, the control of hydraulic systems is
\r\ncomplicated by strong nonlinearities and variability among
\r\nmachines, making the design of effective controllers a sig-
\r\nnificant challenge.
\r\nIn this paper, we propose a data-driven Model Predictive
\r\nControl (MPC) system, initially developed for trajectory
\r\ntracking and subsequently adapted for a path following
\r\napproach. This adaptation is crucial because the trajectory
\r\ntracking method relies on open-loop references, using a
\r\npredefined speed profile that does not account for the
\r\ndynamics of the hydraulic excavator, potentially leading to
\r\ndifficult-to-follow trajectories.
\r\nThe prediction model used in the MPC is based on Linear
\r\nLocal Neuro-Fuzzy Models, trained with the LOcal LInear
\r\nMOdel Tree (LOLIMOT) algorithm, while the linear parame-
\r\nters are refined using the Simulation Error Method (SEM).
\r\nThe proposed control system was rigorously tested on
\r\na JCB Hydradig 110W following a comprehensive data
\r\ncollection campaign to obtain the necessary data for the
\r\ndata-driven model of the hydraulic cylinders.
\r\nResults, evaluated using metrics such as root mean
\r\nsquare error (RMSE) between the actual and reference
\r\npaths, maximum error, and standard deviation (indicating
\r\noscillations during motion), demonstrate that our approach
\r\noutperforms previous data-driven feed-forward controllers,
\r\nhighlighting its efficacy in enhancing hydraulic automation.},
\r\nkeywords = {},
\r\npubstate = {forthcoming},
\r\ntppubtype = {workingpaper}
\r\n}
\r\n<\/pre><\/div>
Close<\/a><\/p><\/div>
Hydraulic excavators are essential construc-
\r\ntion machines widely utilized for ground shaping tasks,
\r\nsuch as horizontal leveling and creating sloped surfaces.
\r\nThese operations require a high degree of precision, which
\r\ncan be challenging for unskilled workers.
\r\nThe implementation of automation in hydraulic machin-
\r\nery has the potential to significantly enhance productivity
\r\nby improving accuracy and reducing reliance on highly
\r\ntrained labor. However, the control of hydraulic systems is
\r\ncomplicated by strong nonlinearities and variability among
\r\nmachines, making the design of effective controllers a sig-
\r\nnificant challenge.
\r\nIn this paper, we propose a data-driven Model Predictive
\r\nControl (MPC) system, initially developed for trajectory
\r\ntracking and subsequently adapted for a path following
\r\napproach. This adaptation is crucial because the trajectory
\r\ntracking method relies on open-loop references, using a
\r\npredefined speed profile that does not account for the
\r\ndynamics of the hydraulic excavator, potentially leading to
\r\ndifficult-to-follow trajectories.
\r\nThe prediction model used in the MPC is based on Linear
\r\nLocal Neuro-Fuzzy Models, trained with the LOcal LInear
\r\nMOdel Tree (LOLIMOT) algorithm, while the linear parame-
\r\nters are refined using the Simulation Error Method (SEM).
\r\nThe proposed control system was rigorously tested on
\r\na JCB Hydradig 110W following a comprehensive data
\r\ncollection campaign to obtain the necessary data for the
\r\ndata-driven model of the hydraulic cylinders.
\r\nResults, evaluated using metrics such as root mean
\r\nsquare error (RMSE) between the actual and reference
\r\npaths, maximum error, and standard deviation (indicating
\r\noscillations during motion), demonstrate that our approach
\r\noutperforms previous data-driven feed-forward controllers,
\r\nhighlighting its efficacy in enhancing hydraulic automation.<\/div>
Close<\/a><\/p><\/div><\/div><\/div>
2.<\/div>
 Msaad, Salim;  Cecchin, Leonardo;  Demir, Ozan;  Fagiano, Lorenzo<\/p>
Data-Driven Model Predictive Control of an Hydraulic Excavator via Local Model Networks Proceedings Article<\/span> Forthcoming<\/span><\/p>
In: <\/span>2025 American Control Conference (ACC), <\/span>Forthcoming.<\/p>
Abstract<\/a><\/span> | BibTeX<\/a><\/span><\/p>
@inproceedings{msaad_data-driven_2025,
\r\ntitle = {Data-Driven Model Predictive Control of an Hydraulic Excavator via Local Model Networks},
\r\nauthor = {Salim Msaad and Leonardo Cecchin and Ozan Demir and Lorenzo Fagiano},
\r\nyear  = {2025},
\r\ndate = {2025-07-01},
\r\nurldate = {2025-07-01},
\r\npublisher = {2025 American Control Conference (ACC)},
\r\nabstract = {A novel solution to control an hydraulic excavator during grading tasks is proposed, featuring a Model Predictive Controller designed using Local Model Networks (LMNs), i.e. linear time-invariant dynamic models averaged by nonlinear static functions. The Local Linear Models Tree (LoLiMoT) algorithm is employed to derive an LMN from experimental data of a real excavator. Then, a nonlinear MPC law is designed and implemented on the excavator\u2019s embedded control system. To further improve the computational efficiency, a time-varying MPC law is designed as well, where the LMN is linearized in real-time using the previously computed optimal trajectory. Experimental results, conducted with the excavator in realworld conditions, show the effectiveness of both approaches in achieving performance comparable to state-of-the-art solutions, while utilizing a more compact dataset and without the need of the hydraulic cylinders\u2019 pressure measurement.},
\r\nkeywords = {},
\r\npubstate = {forthcoming},
\r\ntppubtype = {inproceedings}
\r\n}
\r\n<\/pre><\/div>
Close<\/a><\/p><\/div>
A novel solution to control an hydraulic excavator during grading tasks is proposed, featuring a Model Predictive Controller designed using Local Model Networks (LMNs), i.e. linear time-invariant dynamic models averaged by nonlinear static functions. The Local Linear Models Tree (LoLiMoT) algorithm is employed to derive an LMN from experimental data of a real excavator. Then, a nonlinear MPC law is designed and implemented on the excavator\u2019s embedded control system. To further improve the computational efficiency, a time-varying MPC law is designed as well, where the LMN is linearized in real-time using the previously computed optimal trajectory. Experimental results, conducted with the excavator in realworld conditions, show the effectiveness of both approaches in achieving performance comparable to state-of-the-art solutions, while utilizing a more compact dataset and without the need of the hydraulic cylinders\u2019 pressure measurement.<\/div>
Close<\/a><\/p><\/div><\/div><\/div>
3.<\/div>
 Allamaa, Jean Pierre;  Patrinos, Panagiotis;  Son, Tong Duy<\/p>
ExAMPC: the Data-Driven Explainable and Approximate NMPC with Physical Insights<\/a> Working paper<\/span> <\/p>
2025<\/span>, (Submitted for possible publication in IEEE)<\/span>.<\/p>
Abstract<\/a><\/span> | Links<\/a><\/span> | BibTeX<\/a><\/span><\/p>
@workingpaper{allamaa2025exampc,
\r\ntitle = {ExAMPC: the Data-Driven Explainable and Approximate NMPC with Physical Insights},
\r\nauthor = {Jean Pierre Allamaa and Panagiotis Patrinos and Tong Duy Son},
\r\nurl = {https:\/\/arxiv.org\/abs\/2503.00654},
\r\nyear  = {2025},
\r\ndate = {2025-06-18},
\r\nabstract = {Amidst the surge in the use of Artificial Intelligence (AI) for control purposes, classical and model-based control methods maintain their popularity due to their transparency and deterministic nature. However, advanced controllers like Nonlinear Model Predictive Control (NMPC), despite proven capabilities, face adoption challenges due to their computational complexity and unpredictable closed-loop performance in complex validation systems. This paper introduces ExAMPC, a methodology bridging classical control and explainable AI by augmenting the NMPC with data-driven insights to improve the trustworthiness and reveal the optimization solution and closed-loop performance's sensitivities to physical variables and system parameters. By employing a low-order spline embedding to reduce the open-loop trajectory dimensionality by over 95%, and integrating it with SHAP and Symbolic Regression from eXplainable AI (XAI) for an approximate NMPC, we enable intuitive physical insights into the NMPC's optimization routine. The prediction accuracy of the approximate NMPC is enhanced through physics-inspired continuous-time constraints penalties, reducing the predicted continuous trajectory violations by 93%. ExAMPC enables accurate forecasting of the NMPC's computational requirements with explainable insights on worst-case scenarios. Experimental validation on automated valet parking and autonomous racing with lap-time optimization NMPC, demonstrates the methodology's practical effectiveness in real-world applications.},
\r\nnote = {Submitted for possible publication in IEEE},
\r\nkeywords = {},
\r\npubstate = {published},
\r\ntppubtype = {workingpaper}
\r\n}
\r\n<\/pre><\/div>
Close<\/a><\/p><\/div>
Amidst the surge in the use of Artificial Intelligence (AI) for control purposes, classical and model-based control methods maintain their popularity due to their transparency and deterministic nature. However, advanced controllers like Nonlinear Model Predictive Control (NMPC), despite proven capabilities, face adoption challenges due to their computational complexity and unpredictable closed-loop performance in complex validation systems. This paper introduces ExAMPC, a methodology bridging classical control and explainable AI by augmenting the NMPC with data-driven insights to improve the trustworthiness and reveal the optimization solution and closed-loop performance's sensitivities to physical variables and system parameters. By employing a low-order spline embedding to reduce the open-loop trajectory dimensionality by over 95%, and integrating it with SHAP and Symbolic Regression from eXplainable AI (XAI) for an approximate NMPC, we enable intuitive physical insights into the NMPC's optimization routine. The prediction accuracy of the approximate NMPC is enhanced through physics-inspired continuous-time constraints penalties, reducing the predicted continuous trajectory violations by 93%. ExAMPC enables accurate forecasting of the NMPC's computational requirements with explainable insights on worst-case scenarios. Experimental validation on automated valet parking and autonomous racing with lap-time optimization NMPC, demonstrates the methodology's practical effectiveness in real-world applications.<\/div>
Close<\/a><\/p><\/div>