Interface de controle por métodos de autonomia adaptável deslizante para robôs de inspeção

Abstract

This work implements a human-robot interface for a climbing robot for inspecting weld beads in storage tanks in the petrochemical industry. To accomplish this, first a research on the state-ofthe-art of Human-Robot Interaction, Autonomy, Sliding Autonomy and Levels of Autonomy in robotics systems and autonomous systems was presented. Then, an industrial joystick that works with radio frequency with an transmitter and a receiver, commonly used to move cranes and hoists, was specified and adapted to control the robot. A driver was developed to make this joystick compatible with the Robot Operating System – ROS, used in this work. To add more control and input data from the operator, a electromyographic armband called Myo was appended to the system. This armband is worn in the forearm and is capable of detecting gestures from the operator and rotation angles from the arm. Information from the industrial joystick and the armband are used to control the robot via a Fuzzy controller. This controller works in the sliding autonomy during the robot operation, using as inputs data from the angular velocity of the industrial controller, electromyography reading, weld bead position in the storage tank and rotation angles executed by the operator’s arm, to generate a system capable of recognition of the operator’s skill and correct mistakes from the operator in operating time. The output from the Fuzzy controller is the level of autonomy to be used by the robot. The levels implemented are: Manual, where the operator controls the angular and linear velocities of the robot; Shared, where the velocities are shared between operator and the autonomous system; Supervisory, where the robot controls the angular velocity to stay in the weld bead and the operator controls the linear velocity; and Autonomous, where an end point is defined by the operator and the robot controls both linear and angular velocities. This levels of autonomy along with the proposed sliding autonomy are then analyzed through robot experiments in a simulated environment, showing the purpose of each one of these modes.