The WAM arm : modelling, control and its application in a HMI based on gaze tracking

摘要

In this thesis we describe the design and implementation of a Human Machine\udInterface (HMI) based on gaze tracking proposed to control robot prostheses. Robot\udmanipulators hold a strong similarity with arm prosthetics, we used a 7 degrees\udof freedom (DOF) whole arm manipulator to test our HMI in the execution of\udreaching and grasping tasks. We showed that the interface worked under different\udcontrol strategies using several velocity profiles. The system was tested by ten\udsubjects with encouraging results. We analysed the performance of the 7-DOF\udrobot manipulator in order to determine the suitability of its application in the\uddevelopment of this project. The original setup of the manipulator worked under\udjoint Proportional and Derivative (PD) control but considering the results of the\udinitial analysis of the system we proposed two alternative control strategies aimed\udto improve the performance of the manipulator: a feedforward friction compensation\udtechnique and joint Proportional Integral and Derivative control (PID). We created\uda dynamic model of the 7-DOF manipulator in Simmechanics in order to have a\udbetter understanding of the system. The friction phenomena of the manipulator\udwas identified, represented through a fitted model and included in the system’s\udmodel with the aim of incrementing its accuracy with respect to the real system.\udThe characteristics of the model made it suitable to test and to design control\udstrategies for motion and friction compensation in MATLAB/Simulink. The model\udof the system was validated using data from the real robot arm and it was used\udlater to tune the PID controllers of the joints of the 7-DOF manipulator using\udIterative Feedback Tuning (IFT). Both experimental data and model simulations\udwere used for the tuning procedure considering two different approaches. The data\udobtained from the friction identification process was used to implement a module for\udfeedforward friction compensation over the pre-configured joint PD control of the\udmanipulator. The responses of the system when using joint PID control and joint\udPD control with gravity and friction compensation were compared in the execution\udof motion tasks.