Artigos Acadêmicos e Noticiosos
URI permanente desta comunidadehttps://repositorio.insper.edu.br/handle/11224/3226
Navegar
2 resultados
Resultados da Pesquisa
- Digital Twin data architecture for Product-Service Systems(2024) Durão, Luiz Fernando C.S.; Zancul, Eduardo; Schützer, KlausThe digital representation of physical products by Digital Twins has been increasingly perceived as a relevant enabler for Product-Service Systems (PSS). Digital Twins concentrate product data that can be applied to generate insights and thus support value-added services. However, the literature on the intersection between Digital Twin and PSS has only recently started to receive more attention. There are still research gaps related to the required data and systems integration. In this context, this research aims to propose a Digital Twin data architecture to support Product-Service Systems operation. A Design Science Research (DSR) approach is applied, and the proposed architecture has been implemented and tested. Assessment results indicated that the proposed Digital Twin architecture fulfills the four requirements established from the literature: 1) facilitate and support the service offering; 2) acquire and transmit field operation and customer data; 3) integrate design and manufacturing data; 4) guarantee real-time monitoring, data integration, and data fidelity. The presented results provide an original contribution to the research area and can serve as a reference for applying Digital Twin to support PSS in practice.
- Cable SCARA Robot Controlled by a Neural Network Using Reinforcement Learning(2023) Okabe, Eduardo; Paiva, Victor; Silva-Teixeira, Luis H.; Izuka, JaimeIn this work, three reinforcement learning algorithms (Proximal Policy Optimization, Soft Actor-Critic, and Twin Delayed Deep Deterministic Policy Gradient) are employed to control a two link selective compliance articulated robot arm (SCARA) robot. This robot has three cables attached to its end-effector, which creates a triangular shaped workspace. Positioning the end-effector in the workspace is a relatively simple kinematic problem, but moving outside this region, although possible, requires a nonlinear dynamic model and a state-of-the-art controller. To solve this problem in a simple manner, reinforcement learning algorithms are used to find possible trajectories for three targets out of the workspace. Additionally, the SCARA mechanism offers two possible configurations for each end-effector position. The algorithm results are compared in terms of displacement error, velocity, and standard deviation among ten trajectories provided by the trained network. The results indicate the Proximal Policy Algorithm as the most consistent in the analyzed situations. Still, the Soft Actor-Critic presented better solutions, and Twin Delayed Deep Deterministic Policy Gradient provided interesting and more unusual trajectories.