Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 8
Encontrada(s) 1 página(s)

Investigo. Repositorio Institucional de la Universidade de Vigo
Publication Artículo científico (JournalArticle)

A method for understanding and digitizing manipulation activities using programming by demonstration in robotic applications

  • Malvido Fresnillo, Pablo
  • Vasudevan, Saigopal
  • Mohammed, Wael M.
  • Martinez Lastra, Jose L.
  • Pérez García, José Antonio
Robots are flexible machines, where the flexibility is achieved, mainly, by the re-programming of the robotic system. To fully exploit the potential of robotic systems, an easy, fast, and intuitive programming methodology is desired. By applying such methodology, robots will be open to a wider audience of potential users (i.e. SMEs, etc.) since the need for a robotic expert in charge of programming the robot will not be needed anymore. This paper presents a Programming by Demonstration approach dealing with high-level tasks taking advantage of the ROS standard. The system identifies the different processes associated to a single-arm human manipulation activity and generates an action plan for future interpretation by the robot. The system is composed of five modules, all of them containerized and interconnected by ROS. Three of these modules are in charge of processing the manipulation data gathered by the sensors system, and converting it from the lowest level to the highest manipulation processes. In order to do this transformation, a module is used to train the system. This module generates, for each operation, an Optimized Multiorder Multivariate Markov Model, that later will be used for the operations recognition and process segmentation. Finally, the fifth module is used to interface and calibrate the system. The system was implemented and tested using a dataglove and a hand position tracker to capture the operator’s data during the manipulation. Four users and five different object types were used to train and test the system both for operations recognition and process segmentation and classification, including also the detection of the locations where the operations are performed.
Investigo. Repositorio Institucional de la Universidade de Vigo
Publication Artículo científico (JournalArticle)

An approach based on machine vision for the identification and shape estimation of deformable linear objects

  • Malvido Fresnillo, Pablo
  • Vasudevan, Saigopal
  • Mohammed, Wael M.
  • Martinez Lastra, Jose L.
  • Pérez García, José Antonio
The automation of processes that handle deformable materials, and in particular Deformable Linear Objects (DLOs), such as cables, ropes, and sutures; is a challenging task. Due to their properties, it is very difficult to predict the shape of these objects, making indispensable the use of perception systems for their manipulation. However, the detection of a DLO is a non-trivial task, and it can be even more complicated when additional considerations are made, such as detecting multiple DLOs, with small distances between them or even adjacent to each other, and with occlusions and entanglements between them. In this paper, a novel machine vision approach for estimating the shape of DLOs is proposed to address all these challenges. This approach processes the different DLOs in the image sequentially, repeating the following procedure for each of them. First, the DLO is segmented by examining the colors and edges in the image. Next, the remaining pixels are analyzed using evaluation windows to identify a series of points along the DLO’s skeleton. These points are then employed to model the DLO’s shape using a polynomial function. Finally, the output is evaluated by an unsupervised self-critique module, which validates the results, or fine-tunes the system’s parameters and repeats the process. The performance of the system was tested with several wiring harnesses, detecting all their cables in homogeneous and complex backgrounds, with adjacent cables, and with occlusions. The results show an outstanding performance, with a successful shape estimation rate of more than 90% for some of the system configurations.
Investigo. Repositorio Institucional de la Universidade de Vigo
Publication Artículo científico (JournalArticle)

A dual-arm robotic system for automated multi-branch wire harness assembly in automotive industry

  • Malvido Fresnillo, Pablo
  • Vasudevan, Saigopal
  • Mohammed, Wael M.
  • Pérez García, José Antonio
  • Martinez Lastra, Jose L.
Wire harnesses are critical components in modern vehicles, responsible for transmitting electrical signals and power to sensors and actuators. Despite the high level of automation in the automotive industry, wire harness manufacturing still relies heavilylargely depends on manual assembly. This is due to the significant challenges posed by the process, such as the complexity of perceiving and manipulating flexible materials and the high degree of customization required. As a result, existing solutions only address specific assembly tasks, rather than the entire processare fragmented, unable to scale to full production, and remain economically unviable for high-mix scenarios. To bridge this gap, this paper presents a novel robotic system for fully automating wire harness assembly. The system adopts a task-level programming methodology that leverages process knowledge to enable fast and easy reconfiguration. Additionally, it incorporates specific solutions to address key challenges in multi-branch wire harness manipulation, such as cable separation and entanglement prevention. The system’s performance was evaluated in two real-world assembly scenarios using a dual-arm robot. Experimental results demonstrate the system’s effectiveness and ease of reconfiguration, achieving success rates of 55% and 73% in two complex multi-branch wire harness assembly processes, and highlight areas of improvement, which will be further investigated in future works. The system repository is openly available allowing other researchers to build their solutions upon the proposed methodology.
Investigo. Repositorio Institucional de la Universidade de Vigo
Publication Artículo científico (JournalArticle)

An open and reconfigurable user interface to manage complex ROS-based robotic systems

  • Fresnillo, Pablo Malvido
  • Vasudevan, Saigopal
  • Pérez García, José Antonio
  • Martinez Lastra, Jose L.
The Robot Operating System (ROS) has significantly gained popularity among robotic engineers and researchers over the past five years, primarily due to its powerful infrastructure for node communication, which enables developers to build modular and large robotic applications. However, ROS presents a steep learning curve and lacks the intuitive usability of vendor-specific robotic Graphical User Interfaces (GUIs). Moreover, its modular and distributed nature complicates the control and monitoring of extensive systems, even for advanced users. To address these challenges, this paper introduces a novel, highly adaptable and reconfigurable web-based GUI for intuitively controlling, monitoring, and configuring complex ROS-based robotic systems. The GUI leverages ROSBridge and roslibjs to ensure seamless communication with ROS systems via topics and services. Special attention has been paid to enable the easy reconfiguration of the GUI and to minimize the modifications required to integrate it with new robotic systems. Thus, it has been designed as a versatile platform, which allows for the selective incorporation of modular features to accommodate diverse robotic systems and applications. A predefined template has been defined to ensure the compatibility of the developed features with the GUI platform. Additionally, an initial set of commonly used features in robotic applications is presented. To demonstrate its reconfigurability, the GUI was customized and tested for four industrial use cases, receiving positive feedback. The project’s repository has been made publicly available to support the robotics community and lower the entry barrier for ROS in industrial applications.
Investigo. Repositorio Institucional de la Universidade de Vigo
Publication Artículo científico (JournalArticle)

Generation of realistic synthetic cable images to train deep learning segmentation models

  • MalvidoFresnillo, Pablo
  • Mohammed, Wael M.
  • Vasudevan, Saigopal
  • Pérez García, José Antonio
  • MartinezLastra, Jose L.
Semantic segmentation is one of the most important and studied problems in machine vision, which has been solved with high accuracy by many deep learning models. However, all these models present a significant drawback, they require large and diverse datasets to be trained. Gathering and annotating all these images manually would be extremely time-consuming, hence, numerous researchers have proposed approaches to facilitate or automate the process. Nevertheless, when the objects to be segmented are deformable, such as cables, the automation of this process becomes more challenging, as the dataset needs to represent their high diversity of shapes while keeping a high level of realism, and none of the existing solutions have been able to address it effectively. Therefore, this paper proposes a novel methodology to automatically generate highly realistic synthetic datasets of cables for training deep learning models in image segmentation tasks. This methodology utilizes Blender to create photo-realistic cable scenes and a Python pipeline to introduce random variations and natural deformations. To prove its performance, a dataset composed of 25000 synthetic cable images and their corresponding masks was generated and used to train six popular deep learning segmentation models. These models were then utilized to segment real cable images achieving outstanding results (over 70% IoU and 80% Dice coefficient for all the models). Both the methodology and the generated dataset are publicly available in the project’s repository.
Investigo. Repositorio Institucional de la Universidade de Vigo
Publication Artículo científico (JournalArticle)

Extending the motion planning framework—MoveIt with advanced manipulation functions for industrial applications

  • Malvido Fresnillo, Pablo
  • Vasudevan, Saigopal
  • Mohammed, Wael M.
  • Martinez Lastra, Jose L.
  • Pérez García, José Antonio
MoveIt is the primary software library for motion planning and mobile manipulation in ROS, and it incorporates the latest advances in motion planning, control and perception. However, it is still quite recent, and some important functions to build more advanced manipulation applications, required to robotize many manufacturing processes, have not been developed yet. MoveIt is an open source software, and it relies on the contributions from its community to keep improving and adding new features. Therefore, in this paper, its current state is analyzed to find out which are its main necessities and provide a solution to them. In particular, three gaps of MoveIt are addressed: the automatic tool changing at runtime, the generation of trajectories with full control over the end effector path and speed, and the generation of dual-arm trajectories using different synchronization policies. These functions have been tested with a Motoman SDA10F dual-arm robot, demonstrating their validity in different scenarios. All the developed solutions are generic and robot-agnostic, and they are openly available to be used to extend the capabilities of MoveIt.
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Human to robot hand motion mapping methods: review and classification

  • Meattini, Roberto
  • Suárez Feijóo, Raúl|||0000-0002-3853-7095
  • Palli, Gianluca
  • Melchiorri, Claudio
In this article, the variety of approaches proposed in literature to address the problem of mapping human to robot hand motions are summarized and discussed. We particularly attempt to organize under macro-categories the great quantity of presented methods, that are often difficult to be seen from a general point of view due to different fields of application, specific use of algorithms, terminology and declared goals of the mappings. Firstly, a brief historical overview is reported, in order to provide a look on the emergence of the human to robot hand mapping problem as a both conceptual and analytical challenge that is still open nowadays. Thereafter, the survey mainly focuses on a classification of modern mapping methods under six categories: direct joint, direct Cartesian, taskoriented, dimensionality reduction based, pose recognition based and hybrid mappings. For each of these categories, the general view that associates the related reported studies is provided, and representative references are highlighted. Finally, a concluding discussion along with the authors’ point of view regarding future desirable trends are reported., This work was supported in part by the European Commission’s Horizon 2020 Framework Programme with the project REMODEL under Grant 870133 and in part by the Spanish Government under Grant PID2020-114819GB-I00., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Human to robot hand motion mapping methods: review and classification

  • Meattini, Roberto
  • Suárez Feijóo, Raúl
  • Palli, Gianluca
  • Melchiorri, Claudio
In this article, the variety of approaches proposed in literature to address the problem of mapping human to robot hand motions are summarized and discussed. We particularly attempt to organize under macro-categories the great quantity of presented methods, that are often difficult to be seen from a general point of view due to different fields of application, specific use of algorithms, terminology and declared goals of the mappings. Firstly, a brief historical overview is reported, in order to provide a look on the emergence of the human to robot hand mapping problem as a both conceptual and analytical challenge that is still open nowadays. Thereafter, the survey mainly focuses on a classification of modern mapping methods under six categories: direct joint, direct Cartesian, taskoriented, dimensionality reduction based, pose recognition based and hybrid mappings. For each of these categories, the general view that associates the related reported studies is provided, and representative references are highlighted. Finally, a concluding discussion along with the authors’ point of view regarding future desirable trends are reported., This work was supported in part by the European Commission’s Horizon 2020 Framework Programme with the project REMODEL under Grant 870133 and in part by the Spanish Government under Grant PID2020-114819GB-I00., Peer Reviewed, Postprint (published version)