Dynamic Simulation and Position Control of Parallel Manipulator using Predictive Control Techniques.

• Grant number: 09/05387-5
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: September 01, 2009
• End date: February 29, 2012
• Field of knowledge: Engineering - Mechanical Engineering - Mechanics of Solids
• Principal Investigator: João Mauricio Rosário
• Grantee: Fabian Andres Lara Molina
• Host Institution: Faculdade de Engenharia Mecânica (FEM). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

This proposal addresses the synthesis and application of a control law based on predictive control in order to improve the dynamic performance of parallel kinematics machines (Parallel Kinematic Machines - PKM).The PKM's are compoused by a parallel structure mechanism; this feature provides advantages with respect to serial machines such as better dynamic performance, greater rigidity and load - weight ratio. To further enhance the performance of a PKM is necessary to develop advanced controllers, considering the plant characteristics such as non-linearity of the model and the dynamic response requirements (high speed operation and minimizing the error following the trajectories). The use of predictive controllers offers several advantages to address these problems.The parallel manipulator Orthoglide of three degrees of freedom will be used in this work, in partnership with institutions: SUPELEC and Ecole Centrale de Nantes, in order to validate and compare the predictive controller for the classic controllers. The outcome in this research project is a predictive control law to improve dynamic response of PKM. This control law can be extrapolated to new or existing parallel machines with applications in the domestic industry.To meet this objective, this research project will be divided into following phases: (a) Modeling the Dynamics PKM involving: geometric, kinematic and dynamic model (b) Based on the PKM model, the position controllers will be synthesized for the joint space: classical PID controllers as RST form, dynamic controller and generalized predictive controller ( GPC); (c) Dynamic simulation of position control based on the study kinematic and dynamic model of a PKM in MATLAB-Simulink environment; (d) Implementation and experimental validation of the controller in a robot Orthoglide existing in IRCCyN, Ecole Centrale de Nantes (e) Preparation of technical documents, scientific articles and doctoral thesis at UNICAMP.

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

MOLINA, Fabian Andres Lara. Simulation and experimental implementation of a generalized predictive control for an orthoglide robot based on the dynamic modeling. 2012. Doctoral Thesis - Universidade Estadual de Campinas (UNICAMP). Faculdade de Engenharia Mecânica Campinas, SP.