Embedded visual servoing for planar parallel kinematic machines

• Grant number: 17/11959-8
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2017
• End date: August 31, 2018
• Field of knowledge: Engineering - Electrical Engineering - Electrical, Magnetic and Electronic Measurements, Instrumentation
• Principal Investigator: Maira Martins da Silva
• Grantee: Ricardo Vidal Mota Peixoto
• Host Institution: Escola de Engenharia de São Carlos (EESC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

Due to their kinematic architecture, parallel kinematic manipulators (PKMs) are composed of lighter moving parts. As a result, this architecture is a promising alternative for designing not only robotic manipulators with improved dynamic performance but also energy efficient robotic systems. Classical control strategies for robotic manipulators have to be revisited when dealing with the design of parallel kinematic manipulators. Among these strategies, two strategies should be considered: (I) joint space computed torque control and (II) Cartesian space computed torque control. The first strategy (I) requires the calculation of inverse kinematics of the manipulator since the feedback control is performed by using the actual position of the active joints. The second strategy (II) requires the measurement of the position of the end-effector since because the feedback control strategy is based on this measurement. Thus, although mathematically simpler, this choice imposes a significant technical challenge: the measurement of the end-effector. A planar PKM has been built at EESC-USP (FAPESP 2014/01809-0) and some research has been under investigation regarding strategy (II) (FAPESP 2015/25936-4). These actions are being done in Matlab and DSPACE. Recently, our research group has acquired a computing device, the ODROID-XU4. Using this device, embedded vision computing can be implemented using OpenCV library. The objective of this research project is to explore this device for implementing this strategy (II). Due to the library's versatility and high computing efficiency, it is expected that the Cartesian space computed torque control could achieve improved performance. (AU)