A Visual Servoing Strategy Under Limited Frame Rates for Planar Parallel Kinematic Machines

Autor(es)

Colombo, Fernanda Thais ^[1] ; de Carvalho Fontes, Joao Vitor ^[1] ; da Silva, Maira Martins ^[1]

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Sao Carlos Sch Engn, Dept Mech Engn, Av Trabalhador Saocarlense 400, BR-13566590 Sao Carlos, SP - Brazil

• Tipo de documento: Artigo Científico
• Fonte: JOURNAL OF INTELLIGENT & ROBOTIC SYSTEMS; v. 96, n. 1, p. 95-107, OCT 2019.
• Citações Web of Science: 0

Resumo

In this manuscript, an affordable alternative to control parallel kinematic machines is proposed based in a high-authority/low-authority control strategy. The high-authority loop is implemented using a model-based Cartesian control strategy. This feedback loop exploits a Pose-Based Visual Servo (PBVS) using an eye-to-hand limited frame rate camera. The low-authority loop is implemented using an independent space joint space control strategy. This proposal is experimentally compared with the Cartesian space computed torque control and with the independent joint space control strategies. The experimental campaign is carried out for tracking-trajectory problems using a planar 3R RR parallel manipulator. The experimental results demonstrate that the proposal can achieve results that are more accurate without increasing the manipulator's energy consumption when compared to the use of the independent joint space control strategy. For a predefined square trajectory, a reduction of 4.9% and 24.5% was achieved in the root mean square error of the linear and angular positions, respectively. (AU)

• Processo FAPESP: 14/01809-0 - Alcançando altas acelerações com manipuladores robóticos planares, Fase I: redundância cinemática
• Beneficiário: Maira Martins da Silva
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 15/25936-4 - Torque Calculado no Espaço da Tarefa em Manipuladores Paralelos Planos utilizando Servovisão
• Beneficiário: Fernanda Thaís Colombo
• Modalidade de apoio: Bolsas no Brasil - Mestrado