Pedro Ferreira
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Pedro Ferreira has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
- Niels Lohse
The Birmingham Institute for Robotics, College of Engineering and Physical Sciences, School of Engineering, University of Birmingham, Birmingham, U.K.
7 shared publications - Laura Justham
Wolfson School of Mechanical, Electrical, and Manufacturing Engineering, Loughborough University, Loughborough LE11 3TU, UK
4 shared publications - Antonio Maffei
KTH Royal Institute of Technology
3 shared publications - Francesco Lupi
Department of Civil and Industrial Engineering, University of Pisa, 56122 Pisa, Italy
3 shared publications - Matthew Beatty
Wolfson School of Mechanical, Electrical, and Manufacturing Engineering, Loughborough University, Loughborough LE11 3TU, UK
3 shared publications - Michele Lanzetta
Department of Civil and Industrial Engineering, University of Pisa, Pisa, Italy
3 shared publications
Latest Publications
- Open Access
- 3 November 2026
- Published by Elsevier in Journal of Manufacturing Systems
- Vol. 89, 1077-1106
Human behaviour and intention prediction for industrial human-robot collaboration: Potentials, challenges, and opportunities
- 3 November 2026
- Published by Elsevier in Journal of Manufacturing Systems
- Vol. 89, 1077-1106
Open AccessHuman behaviour and intention prediction for industrial human-robot collaboration: Potentials, challenges, and opportunities
- Open Access
- 28 September 2026
- Published by Springer Nature
- p. 255-269
AI-Enhanced Serious Games in Production Management: An Exploratory Review of Emerging Applications
This paper examines the integration of artificial intelligence (AI) and serious games for training and decision support in production management. As manufacturing systems become increasingly complex under Industry 4.0 and 5.0 paradigms, traditional instructional approaches struggle to develop adaptive decision-making skills. Serious games provide interactive, risk-free environments, while AI enables adaptive learning, intelligent feedback, and dynamic scenario generation. A structured exploratory review was conducted to analyse recent contributions in this domain. The findings identify three main typologies: AI-driven virtual players, machine learning-based preference capture and scheduling optimisation systems, and simulations enhanced with intelligent feedback. Results consistently highlight increased motivation and engagement, while learning outcomes are generally limited to short-term effects, with no evidence of long-term retention or workplace transfer. The review introduces a dual-purpose perspective, positioning AI-enhanced serious games as both training tools and instruments for knowledge elicitation in operations research. Key challenges include limited empirical validation, small sample sizes, high development costs, cognitive load issues, and ethical concerns. The study provides a structured foundation for future research and emphasises the need for longitudinal validation in industrial contexts.
- 28 September 2026
- Published by Springer Nature
- p. 255-269
Open AccessAI-Enhanced Serious Games in Production Management: An Exploratory Review of Emerging Applications
This paper examines the integration of artificial intelligence (AI) and serious games for training and decision support in production management. As manufacturing systems become increasingly complex under Industry 4.0 and 5.0 paradigms, traditional instructional approaches struggle to develop adaptive decision-making skills. Serious games provide interactive, risk-free environments, while AI enables adaptive learning, intelligent feedback, and dynamic scenario generation. A structured exploratory review was conducted to analyse recent contributions in this domain. The findings identify three main typologies: AI-driven virtual players, machine learning-based preference capture and scheduling optimisation systems, and simulations enhanced with intelligent feedback. Results consistently highlight increased motivation and engagement, while learning outcomes are generally limited to short-term effects, with no evidence of long-term retention or workplace transfer. The review introduces a dual-purpose perspective, positioning AI-enhanced serious games as both training tools and instruments for knowledge elicitation in operations research. Key challenges include limited empirical validation, small sample sizes, high development costs, cognitive load issues, and ethical concerns. The study provides a structured foundation for future research and emphasises the need for longitudinal validation in industrial contexts.
- Open Access
Integrating Sustainable Development Goals into a Practically Applicable Sustainable Value Stream Mapping
As manufacturing organisations work to align their operations with global sustainability expectations, the integration of the United Nations (UN) Sustainable Development Goals (SDGs) into production systems has become increasingly important and a critical pathway toward Industry 5.0. Sustainable Value Stream Mapping (Sus-VSM) provides a structured approach for assessing economic, environmental, and social performance, but its practical adoption remains limited, and it has not been systematically aligned with SDG Targets. These limitations are particularly evident among small and medium-sized enterprises (SMEs) which often lack the resources needed to implement extensive sustainability indicator sets. This study combines a systematic literature review with industrial evidence to identify Sus-VSM indicators and examine their use in practice. A consolidated set of 18 economic, 22 environmental, and 18 social indicators is derived and mapped to 16 SDG Targets relevant to manufacturing by means of a Delphi-based expert assessment. The mapping results are analysed to evaluate indicator usefulness and expert agreement and are complemented by an industrial verification from 30 companies across different sectors that rates each indicator in terms of relevance to SDGs and applicability in real factory contexts. The results show that economic indicators are most aligned with SDG 8, environmental indicators with SDG 9, SDG 12, and SDG 13, and social indicators with SDG 3 and SDG 8, while gaps persist for water, energy transition, and some climate-related Targets. A prioritised indicator set is proposed that maintains coverage of the selected SDG Targets and remains feasible for SMEs, providing a practically oriented basis for embedding SDG-aligned sustainability assessment within Sus-VSM.
Open AccessIntegrating Sustainable Development Goals into a Practically Applicable Sustainable Value Stream Mapping
As manufacturing organisations work to align their operations with global sustainability expectations, the integration of the United Nations (UN) Sustainable Development Goals (SDGs) into production systems has become increasingly important and a critical pathway toward Industry 5.0. Sustainable Value Stream Mapping (Sus-VSM) provides a structured approach for assessing economic, environmental, and social performance, but its practical adoption remains limited, and it has not been systematically aligned with SDG Targets. These limitations are particularly evident among small and medium-sized enterprises (SMEs) which often lack the resources needed to implement extensive sustainability indicator sets. This study combines a systematic literature review with industrial evidence to identify Sus-VSM indicators and examine their use in practice. A consolidated set of 18 economic, 22 environmental, and 18 social indicators is derived and mapped to 16 SDG Targets relevant to manufacturing by means of a Delphi-based expert assessment. The mapping results are analysed to evaluate indicator usefulness and expert agreement and are complemented by an industrial verification from 30 companies across different sectors that rates each indicator in terms of relevance to SDGs and applicability in real factory contexts. The results show that economic indicators are most aligned with SDG 8, environmental indicators with SDG 9, SDG 12, and SDG 13, and social indicators with SDG 3 and SDG 8, while gaps persist for water, energy transition, and some climate-related Targets. A prioritised indicator set is proposed that maintains coverage of the selected SDG Targets and remains feasible for SMEs, providing a practically oriented basis for embedding SDG-aligned sustainability assessment within Sus-VSM.
- Open Access
A Methodological Framework for Incremental Capacity-Based Feature Engineering and Unsupervised Learning Across First-Life and Second-Life Battery Datasets
Accurately assessing battery health across mixed datasets remains a challenge due to differences in chemistry, format, and usage history. This study presents a reproducible framework for preparing battery cycling data using incremental capacity analysis (ICA), with the aim of supporting machine learning (ML) workflows across both first-life and second-life battery datasets. The methodology includes IC curve generation, feature extraction, encoding and scaling, feature reduction, and unsupervised learning exploration. A two-tiered outlier detection system was introduced during preprocessing to flag edge-case samples. Two clustering algorithms, K-means and HDBSCAN, were applied to the engineered feature space to explore patterns in the IC feature space. K-means revealed broad health-related groupings with overlapping boundaries, while HDBSCAN identified finer clusters and flagged additional ambiguous samples as noise. To support interpretation, PCA and t-SNE were used to visualise the feature space in reduced dimensions. Rather than using clustering as a classification tool, the resulting cluster and noise labels are proposed as structure-aware meta-features for supervised learning. The framework accommodates heterogeneous battery datasets and addresses the challenges of integrating data from mixed sources with varying histories and characteristics. These outputs provide a structured foundation for future supervised classification of battery state of health.
Open AccessA Methodological Framework for Incremental Capacity-Based Feature Engineering and Unsupervised Learning Across First-Life and Second-Life Battery Datasets
Accurately assessing battery health across mixed datasets remains a challenge due to differences in chemistry, format, and usage history. This study presents a reproducible framework for preparing battery cycling data using incremental capacity analysis (ICA), with the aim of supporting machine learning (ML) workflows across both first-life and second-life battery datasets. The methodology includes IC curve generation, feature extraction, encoding and scaling, feature reduction, and unsupervised learning exploration. A two-tiered outlier detection system was introduced during preprocessing to flag edge-case samples. Two clustering algorithms, K-means and HDBSCAN, were applied to the engineered feature space to explore patterns in the IC feature space. K-means revealed broad health-related groupings with overlapping boundaries, while HDBSCAN identified finer clusters and flagged additional ambiguous samples as noise. To support interpretation, PCA and t-SNE were used to visualise the feature space in reduced dimensions. Rather than using clustering as a classification tool, the resulting cluster and noise labels are proposed as structure-aware meta-features for supervised learning. The framework accommodates heterogeneous battery datasets and addresses the challenges of integrating data from mixed sources with varying histories and characteristics. These outputs provide a structured foundation for future supervised classification of battery state of health.
- Open Access
- 23 April 2025
- Published by MDPI AG in Engineering Proceedings
- Vol. 82 (1) , 111
Real-Time Hardness Prediction Using COTS Tactile Sensors in Robotic Grippers
In material classification tasks, custom sensors have traditionally been employed to achieve accuracy scores. While numerous studies have reported high accuracy rates, there has been limited discussion on real-time predictions or real-world applications in most research papers. The real-time prediction of object material properties is crucial for enhancing the tactile sensing capabilities of robotics in industrial settings. This study proposes the use of Commercial Off-The-Shelf (COTS) tactile sensors for hardness classification, utilizing small datasets for model training and real-time prediction. Testing involves evaluating the ability of robotic grippers to accurately predict the hardness of new, unknown objects, categorizing them into two classes (soft, hard) or three classes (hard, soft, flexible). Results obtained from a multiple-algorithm approach reveal an 80% accuracy rate for binary classification, with real-time tests demonstrating two out of three correct predictions for most sensors. For ternary classification, the accuracy rate is around 70%, with two out of three correct predictions from at least one sensor. These findings highlight the capability of COTS sensors to perform real-time hardness classification effectively. This also highlights that COTS sensors have capabilities and flexibility based on their dimensional architecture that can be used in many different robotics applications without investing time in the development of a specific use-case sensor for classification tasks within robotic tactile sensing.
- 23 April 2025
- Published by MDPI AG in Engineering Proceedings
- Vol. 82 (1) , 111
Open AccessReal-Time Hardness Prediction Using COTS Tactile Sensors in Robotic Grippers
In material classification tasks, custom sensors have traditionally been employed to achieve accuracy scores. While numerous studies have reported high accuracy rates, there has been limited discussion on real-time predictions or real-world applications in most research papers. The real-time prediction of object material properties is crucial for enhancing the tactile sensing capabilities of robotics in industrial settings. This study proposes the use of Commercial Off-The-Shelf (COTS) tactile sensors for hardness classification, utilizing small datasets for model training and real-time prediction. Testing involves evaluating the ability of robotic grippers to accurately predict the hardness of new, unknown objects, categorizing them into two classes (soft, hard) or three classes (hard, soft, flexible). Results obtained from a multiple-algorithm approach reveal an 80% accuracy rate for binary classification, with real-time tests demonstrating two out of three correct predictions for most sensors. For ternary classification, the accuracy rate is around 70%, with two out of three correct predictions from at least one sensor. These findings highlight the capability of COTS sensors to perform real-time hardness classification effectively. This also highlights that COTS sensors have capabilities and flexibility based on their dimensional architecture that can be used in many different robotics applications without investing time in the development of a specific use-case sensor for classification tasks within robotic tactile sensing.
- Open Access
- 9 February 2025
- Published by MDPI AG in Electronics
- Vol. 14 (4) , 674
Mechanoreceptor-Inspired Tactile Sensor Topological Configurations for Hardness Classification in Robotic Grippers
Human hands have the unique ability to classify material properties, such as hardness, using mechanoreceptors and tactile information. Previous studies have demonstrated hardness classification using Commercial Off-The-Shelf (COTS) sensors but lacked robotic integration considerations. This study explores the integration of multiple COTS sensors, inspired by mechanoreceptors, for classifying material hardness. The sensors were used to classify objects into three categories—hard, soft, and flexible—based on the qualitative Shore hardness scale. The aim was to identify the optimal sensor topology configuration that delivers high accuracy, using machine learning algorithms provided in the literature. The results suggest that the Random Forest Classifier is the most suitable algorithm, showcasing accuracies ranging from 90% to 98.7%, across various sensor topologies. The ‘PFV’ topology, comprising a potentiometer (P), force sensor (F), and vibration sensor (V), achieved the highest accuracy of 98.7%, while the ‘FPV’ and ‘FVP’ recorded accuracies between 96% and 97.5%. The topology of FPV and FVP have the most closely related configuration to that of mechanoreceptors; however, the results show that PFV outperforms this configuration. While the PFV topology marginally outperforms the mechanoreceptor-inspired configurations, the results demonstrate that bio-inspired sensor arrangements provide a robust solution for hardness classification in robotics. The PFV topology performs better than FPV in terms of prediction speed, with an average prediction time of 8.31 ms (millisecond) for PFV versus 13.93 ms for FPV. PFV and FPV achieved 12 and 13 correct predictions, respectively, out of 18 objects. The faster prediction times of PFV make it particularly advantageous for applications requiring quick and accurate decision-making for robotic applications.
- 9 February 2025
- Published by MDPI AG in Electronics
- Vol. 14 (4) , 674
Open AccessMechanoreceptor-Inspired Tactile Sensor Topological Configurations for Hardness Classification in Robotic Grippers
Human hands have the unique ability to classify material properties, such as hardness, using mechanoreceptors and tactile information. Previous studies have demonstrated hardness classification using Commercial Off-The-Shelf (COTS) sensors but lacked robotic integration considerations. This study explores the integration of multiple COTS sensors, inspired by mechanoreceptors, for classifying material hardness. The sensors were used to classify objects into three categories—hard, soft, and flexible—based on the qualitative Shore hardness scale. The aim was to identify the optimal sensor topology configuration that delivers high accuracy, using machine learning algorithms provided in the literature. The results suggest that the Random Forest Classifier is the most suitable algorithm, showcasing accuracies ranging from 90% to 98.7%, across various sensor topologies. The ‘PFV’ topology, comprising a potentiometer (P), force sensor (F), and vibration sensor (V), achieved the highest accuracy of 98.7%, while the ‘FPV’ and ‘FVP’ recorded accuracies between 96% and 97.5%. The topology of FPV and FVP have the most closely related configuration to that of mechanoreceptors; however, the results show that PFV outperforms this configuration. While the PFV topology marginally outperforms the mechanoreceptor-inspired configurations, the results demonstrate that bio-inspired sensor arrangements provide a robust solution for hardness classification in robotics. The PFV topology performs better than FPV in terms of prediction speed, with an average prediction time of 8.31 ms (millisecond) for PFV versus 13.93 ms for FPV. PFV and FPV achieved 12 and 13 correct predictions, respectively, out of 18 objects. The faster prediction times of PFV make it particularly advantageous for applications requiring quick and accurate decision-making for robotic applications.
- Open Access
A Review of Methods of Generating Incremental Capacity–Differential Voltage Curves for Battery Health Determination
Incremental capacity–differential voltage is a powerful tool for transforming raw voltage data from battery cycling data into curves with distinguishable peaks and valleys. These peaks and valleys have been claimed as useful health features in the literature for providing non-invasive, comprehensive insights into a battery’s health and age. Although extensive studies exist on this topic, no standardized approach for generating these curves has been established. This paper analyzes various calculation methodologies and different post-processing filters employed in the literature. These methods are validated using three datasets: two publicly available datasets from Oxford University and a publication from Nature, along with a dataset collected from Loughborough University. The findings highlight the effectiveness of specific calculation methodologies and filters through the differences in the curves produced. Based on the results and analysis, a recommended operational procedure for generating incremental capacity curves is proposed. This standardized procedure aims to enhance the reliability and consistency of producing incremental capacity curves for state-of-health assessments for batteries.
Open AccessA Review of Methods of Generating Incremental Capacity–Differential Voltage Curves for Battery Health Determination
Incremental capacity–differential voltage is a powerful tool for transforming raw voltage data from battery cycling data into curves with distinguishable peaks and valleys. These peaks and valleys have been claimed as useful health features in the literature for providing non-invasive, comprehensive insights into a battery’s health and age. Although extensive studies exist on this topic, no standardized approach for generating these curves has been established. This paper analyzes various calculation methodologies and different post-processing filters employed in the literature. These methods are validated using three datasets: two publicly available datasets from Oxford University and a publication from Nature, along with a dataset collected from Loughborough University. The findings highlight the effectiveness of specific calculation methodologies and filters through the differences in the curves produced. Based on the results and analysis, a recommended operational procedure for generating incremental capacity curves is proposed. This standardized procedure aims to enhance the reliability and consistency of producing incremental capacity curves for state-of-health assessments for batteries.
- 24 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
A Survey on Trends in Visual Inspection Systems Toward Industry 5.0 – A Systematic Mapping Study
Over the past decades, significant developments have occurred in the manufacturing domain, culminating in the contemporary landscape of Industry 5.0 (I5.0). This era is defined by three primary dimensions: technical aspects, sustainability concerns, and human-centricity. Although the broader manufacturing field features a dispersed array of scientific literature reviews amidst this multidisciplinary transition, a focused examination of Visual Inspection Systems (VIS) within this context appears absent. This study aims to provide a comprehensive mapping study that navigates this complex and emerging area of research for VIS, incorporating both scientific publications and Intellectual Property (IP). Employing a systematic methodology for mapping review, an initial exploratory search was conducted, retrieving 264 documents. Following a systematic screening process, 46 documents were identified as relevant, underscoring the preliminary findings and prevailing trends in the field. Results shows that three main clusters emerged with rising interest in technological solutions. Additionally, this study offers a procedural approach to further investigate these promising exploratory findings and expand upon the current review.
- 24 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-9
A Survey on Trends in Visual Inspection Systems Toward Industry 5.0 – A Systematic Mapping Study
Over the past decades, significant developments have occurred in the manufacturing domain, culminating in the contemporary landscape of Industry 5.0 (I5.0). This era is defined by three primary dimensions: technical aspects, sustainability concerns, and human-centricity. Although the broader manufacturing field features a dispersed array of scientific literature reviews amidst this multidisciplinary transition, a focused examination of Visual Inspection Systems (VIS) within this context appears absent. This study aims to provide a comprehensive mapping study that navigates this complex and emerging area of research for VIS, incorporating both scientific publications and Intellectual Property (IP). Employing a systematic methodology for mapping review, an initial exploratory search was conducted, retrieving 264 documents. Following a systematic screening process, 46 documents were identified as relevant, underscoring the preliminary findings and prevailing trends in the field. Results shows that three main clusters emerged with rising interest in technological solutions. Additionally, this study offers a procedural approach to further investigate these promising exploratory findings and expand upon the current review.
- Open Access
- 22 June 2024
- Published by MDPI AG in Electronics
- Vol. 13 (13) , 2450
Hardness Classification Using Cost-Effective Off-the-Shelf Tactile Sensors Inspired by Mechanoreceptors
Perception is essential for robotic systems, enabling effective interaction with their surroundings through actions such as grasping and touching. Traditionally, this has relied on integrating various sensor systems, including tactile sensors, cameras, and acoustic sensors. This study leverages commercially available tactile sensors for hardness classification, drawing inspiration from the functionality of human mechanoreceptors in recognizing complex object properties during grasping tasks. Unlike previous research using customized sensors, this study focuses on cost-effective, easy-to-install, and readily deployable sensors. The approach employs a qualitative method, using Shore hardness taxonomy to select objects and evaluate the performance of commercial off-the-shelf (COTS) sensors. The analysis includes data from both individual sensors and their combinations analysed using multiple machine learning approaches, and accuracy as the primary evaluation metric was considered. The findings illustrate that increasing the number of classification classes impacts accuracy, achieving 92% in binary classification, 82% in ternary, and 80% in quaternary scenarios. Notably, the performance of commercially available tactile sensors is comparable to those reported in the literature, which range from 50% to 98% accuracy, achieving 92% accuracy with a limited data set. These results highlight the capability of COTS tactile sensors in hardness classification giving accuracy levels of 92%, while being cost-effective and easier to deploy than customized tactile sensors.
- 22 June 2024
- Published by MDPI AG in Electronics
- Vol. 13 (13) , 2450
Open AccessHardness Classification Using Cost-Effective Off-the-Shelf Tactile Sensors Inspired by Mechanoreceptors
Perception is essential for robotic systems, enabling effective interaction with their surroundings through actions such as grasping and touching. Traditionally, this has relied on integrating various sensor systems, including tactile sensors, cameras, and acoustic sensors. This study leverages commercially available tactile sensors for hardness classification, drawing inspiration from the functionality of human mechanoreceptors in recognizing complex object properties during grasping tasks. Unlike previous research using customized sensors, this study focuses on cost-effective, easy-to-install, and readily deployable sensors. The approach employs a qualitative method, using Shore hardness taxonomy to select objects and evaluate the performance of commercial off-the-shelf (COTS) sensors. The analysis includes data from both individual sensors and their combinations analysed using multiple machine learning approaches, and accuracy as the primary evaluation metric was considered. The findings illustrate that increasing the number of classification classes impacts accuracy, achieving 92% in binary classification, 82% in ternary, and 80% in quaternary scenarios. Notably, the performance of commercially available tactile sensors is comparable to those reported in the literature, which range from 50% to 98% accuracy, achieving 92% accuracy with a limited data set. These results highlight the capability of COTS tactile sensors in hardness classification giving accuracy levels of 92%, while being cost-effective and easier to deploy than customized tactile sensors.
- 8 September 2023
- Published by Springer Nature
- p. 343-355
Mobile Robots for Collaborative Manipulation over Uneven Ground Using Decentralised Impedance Control
- 8 September 2023
- Published by Springer Nature
- p. 343-355
Mobile Robots for Collaborative Manipulation over Uneven Ground Using Decentralised Impedance Control