Gustavo Barros
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Gustavo Barros has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Bernd Kuhlenkötter
Chair of Production Systems, Ruhr-University Bochum, Bochum, Germany
2 shared publications - Aleksey Gennadievich Aksenov
Institute for Computer Aided Design, Russian Academy of Sciences, Moscow, Russia
1 shared publication - Amedeo Tornambe
INAF -Osservatorio Astronomico di Roma Via di Frascati 33 00040 Monteporzio Catone Italy
1 shared publication Bernd Kuhlenkötter
1 shared publication- Carlo Luciano Bianco
ICRANet
1 shared publication - Enzo Brocato
25INAF - Osservatorio Astronomico d’Abruzzo, Teramo, Italy
1 shared publication
Latest Publications
- Open Access
- 30 June 2026
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 145 (5-6) , 1-21
Data-driven trajectory performance prediction for industrial robots via multi-modal retrieval
Evaluating industrial robot performance typically involves costly individual tests or standardized procedures that cover only a limited set of motion sequences. This makes it difficult to transfer measured accuracy to new trajectories, especially when robot configurations or workspace regions change. To enable data-driven performance prediction at scale, this work introduces a publicly available database of over 30,000 measured trajectories of an ABB IRB4400 industrial robot. Based on this database, a multi-modal retrieval framework is proposed that identifies similar motion sequences and predicts trajectory performance from their measured deviations. The framework employs a two-stage pipeline: embedding-based candidate retrieval across multiple motion modalities followed by constrained Dynamic Time Warping refinement for precise spatio-temporal alignment. Experimental evaluation on 500 randomly sampled trajectories demonstrates that multi-modal retrieval outperforms a scalar-feature nearest-neighbor baseline and a supervised neural-network baseline. Segment-level decomposition further improves both prediction accuracy and computational efficiency. The framework operates without parametric model fitting and can incorporate additional measurements without structural modification, allowing predictive capability to grow with increasing database coverage.
- 30 June 2026
- Published by Springer Nature in The International Journal of Advanced Manufacturing Technology
- Vol. 145 (5-6) , 1-21
Open AccessData-driven trajectory performance prediction for industrial robots via multi-modal retrieval
Evaluating industrial robot performance typically involves costly individual tests or standardized procedures that cover only a limited set of motion sequences. This makes it difficult to transfer measured accuracy to new trajectories, especially when robot configurations or workspace regions change. To enable data-driven performance prediction at scale, this work introduces a publicly available database of over 30,000 measured trajectories of an ABB IRB4400 industrial robot. Based on this database, a multi-modal retrieval framework is proposed that identifies similar motion sequences and predicts trajectory performance from their measured deviations. The framework employs a two-stage pipeline: embedding-based candidate retrieval across multiple motion modalities followed by constrained Dynamic Time Warping refinement for precise spatio-temporal alignment. Experimental evaluation on 500 randomly sampled trajectories demonstrates that multi-modal retrieval outperforms a scalar-feature nearest-neighbor baseline and a supervised neural-network baseline. Segment-level decomposition further improves both prediction accuracy and computational efficiency. The framework operates without parametric model fitting and can incorporate additional measurements without structural modification, allowing predictive capability to grow with increasing database coverage.
- Open Access
- 30 April 2026
- Published by Springer Nature in Research Square
Data-driven trajectory performance prediction for industrial robots via multi-modal retrieval
Evaluating industrial robot performance typically involves costly individual tests or standardized procedures that cover only a limited set of motion sequences. This makes it difficult to transfer measured accuracy to new trajectories, especially when robot configurations or workspace regions change. To enable data-driven performance prediction at scale, this work introduces a publicly available database of over 30,000 measured trajectories of an ABB IRB4400 industrial robot. Based on this database, a multi-modal retrieval framework is proposed that identifies similar motion sequences and predicts trajectory performance from their measured deviations. The framework employs a two-stage pipeline: embedding-based candidate retrieval across multiple motion modalities followed by constrained Dynamic Time Warping refinement for precise spatio-temporal alignment. Experimental evaluation on 500 randomly sampled trajectories demonstrates that multi-modal retrieval outperforms a scalar-feature nearest-neighbor baseline, and that segment-level decomposition improves both prediction accuracy and computational efficiency. The framework operates without parametric model fitting and can incorporate additional measurements without structural modification, allowing predictive capability to grow with increasing database coverage.
- 30 April 2026
- Published by Springer Nature in Research Square
Open AccessData-driven trajectory performance prediction for industrial robots via multi-modal retrieval
Evaluating industrial robot performance typically involves costly individual tests or standardized procedures that cover only a limited set of motion sequences. This makes it difficult to transfer measured accuracy to new trajectories, especially when robot configurations or workspace regions change. To enable data-driven performance prediction at scale, this work introduces a publicly available database of over 30,000 measured trajectories of an ABB IRB4400 industrial robot. Based on this database, a multi-modal retrieval framework is proposed that identifies similar motion sequences and predicts trajectory performance from their measured deviations. The framework employs a two-stage pipeline: embedding-based candidate retrieval across multiple motion modalities followed by constrained Dynamic Time Warping refinement for precise spatio-temporal alignment. Experimental evaluation on 500 randomly sampled trajectories demonstrates that multi-modal retrieval outperforms a scalar-feature nearest-neighbor baseline, and that segment-level decomposition improves both prediction accuracy and computational efficiency. The framework operates without parametric model fitting and can incorporate additional measurements without structural modification, allowing predictive capability to grow with increasing database coverage.
- Open Access
- 20 November 2024
- Published by Springer Nature in Production Engineering
- Vol. 19 (3-4) , 429-453
Performance measurement of unmanned aerial vehicles to suit industrial applications
For the measurement and technical evaluation of industrial robots, a test procedure defined in ISO 9283 has existed for 25 years, with which the performance criteria of industrial robots relevant to users, e.g., path accuracy, pose accuracy, or position overshoot, can be measured. The specified test procedure and defined operating and environmental conditions are to be adhered to enable comparability of the tested robotic systems. In the production environment, unmanned aerial vehicles (UAVs) are increasingly being utilized for tasks such as material transportation and surveillance. However, for these applications, which are currently the subject of considerable discussion, UAVs must also adhere to industrial standards regarding performance criteria. To ensure that these performance criteria are met across all manufacturers, a defined test procedure for UAVs is required in which all relevant performance criteria can be determined based on ISO 9283. This article will, therefore, first conduct the test procedure for a UAV under ISO 9283 and examine the transferability of the standard for performance measurement of UAVs in detail. However, evaluating the measurements suggests that the procedure defined in ISO 9283 is not fully transferable to UAVs as measurements do not fully cover the potential strengths or weaknesses of UAVs regarding the test procedure. The article concludes by discussing modifications to the test procedure and thus suggesting establishing a standardized procedure for UAVs. This procedure should enable the acquisition of performance criteria and might function as a safety benchmark before operating UAVs.
- 20 November 2024
- Published by Springer Nature in Production Engineering
- Vol. 19 (3-4) , 429-453
Open AccessPerformance measurement of unmanned aerial vehicles to suit industrial applications
For the measurement and technical evaluation of industrial robots, a test procedure defined in ISO 9283 has existed for 25 years, with which the performance criteria of industrial robots relevant to users, e.g., path accuracy, pose accuracy, or position overshoot, can be measured. The specified test procedure and defined operating and environmental conditions are to be adhered to enable comparability of the tested robotic systems. In the production environment, unmanned aerial vehicles (UAVs) are increasingly being utilized for tasks such as material transportation and surveillance. However, for these applications, which are currently the subject of considerable discussion, UAVs must also adhere to industrial standards regarding performance criteria. To ensure that these performance criteria are met across all manufacturers, a defined test procedure for UAVs is required in which all relevant performance criteria can be determined based on ISO 9283. This article will, therefore, first conduct the test procedure for a UAV under ISO 9283 and examine the transferability of the standard for performance measurement of UAVs in detail. However, evaluating the measurements suggests that the procedure defined in ISO 9283 is not fully transferable to UAVs as measurements do not fully cover the potential strengths or weaknesses of UAVs regarding the test procedure. The article concludes by discussing modifications to the test procedure and thus suggesting establishing a standardized procedure for UAVs. This procedure should enable the acquisition of performance criteria and might function as a safety benchmark before operating UAVs.
- 1 January 2009
- Published by AIP Publishing
- Vol. 1111 (1) , 325-332
The canonical Gamma-Ray Bursts: long, “fake”-“disguised” and “genuine” short bursts
The Gamma‐Ray Bursts (GRBs) offer the unprecedented opportunity to observe for the first time the blackholic energy extracted by the vacuum polarization during the process of gravitational collapse to a black hole leading to the formation of an electron‐positron plasma. The uniqueness of the Kerr‐Newman black hole implies that very different processes originating from the gravitational collapse a) of a single star in a binary system induced by the companion, or b) of two neutron stars, or c) of a neutron star and a white dwarf, do lead to the same structure for the observed GRB. The recent progress of the numerical integration of the relativistic Boltzmann equations with collision integrals including 2‐body and 3‐body interactions between the particles offer a powerful conceptual tool in order to differentiate the traditional “fireball” picture, an expanding hot cavity considered by Cavallo and Rees, as opposed to the “fireshell” model, composed of an internally cold shell of relativistically expanding electron‐positron‐baryonplasma. The analysis of the fireshell naturally leads to a canonical GRB composed of a proper‐GRB and an extended afterglow. By recalling the three interpretational paradigms for GRBs we show how the fireshell model leads to an understanding of the GRB structure and to an alternative classification of short and long GRBs.
- 1 January 2009
- Published by AIP Publishing
- Vol. 1111 (1) , 325-332
The canonical Gamma-Ray Bursts: long, “fake”-“disguised” and “genuine” short bursts
The Gamma‐Ray Bursts (GRBs) offer the unprecedented opportunity to observe for the first time the blackholic energy extracted by the vacuum polarization during the process of gravitational collapse to a black hole leading to the formation of an electron‐positron plasma. The uniqueness of the Kerr‐Newman black hole implies that very different processes originating from the gravitational collapse a) of a single star in a binary system induced by the companion, or b) of two neutron stars, or c) of a neutron star and a white dwarf, do lead to the same structure for the observed GRB. The recent progress of the numerical integration of the relativistic Boltzmann equations with collision integrals including 2‐body and 3‐body interactions between the particles offer a powerful conceptual tool in order to differentiate the traditional “fireball” picture, an expanding hot cavity considered by Cavallo and Rees, as opposed to the “fireshell” model, composed of an internally cold shell of relativistically expanding electron‐positron‐baryonplasma. The analysis of the fireshell naturally leads to a canonical GRB composed of a proper‐GRB and an extended afterglow. By recalling the three interpretational paradigms for GRBs we show how the fireshell model leads to an understanding of the GRB structure and to an alternative classification of short and long GRBs.