Conferencia del profesor Rolf Johansson, del Departamento de Control Automático de la Universidad de Lund impartirá la conferencia:
Industrial Robots, Skills and Work-Space Sensing
Exteroception or robotic work-space sensing is necessary for efficient usage of robots for purposes of manipulation and manufacturing in new and precision-demanding applications—e.g., robotic precision machining, assembly operations, surgical robotics. In particular, the need for sensor networks and distributed sensing poses new scientific and technological challenges. In addition to established measurement technology for geometric and kinematic data such as robot vision and navigation, there is a need for sensor capacity for physical quantities other than geometric or kinematic ones. In this lecture, we review examples of robotic work-space sensing and control and the current needs for new sensors measuring force, touch, texture, speed, and tool impact. Several robotic assembly use cases would benefit from such new sensor technology. Also, we will compare conditions for sensor-rich and sensor-deprived robotic work spaces. A few case studies of industrial robotics are provided. The role and formalization of robotic skills are discussed.
One way to explore the benefits of exteroception is to start from sensor-deprived robotic work-space control. The traditional way of controlling an industrial robot is to program it to follow desired trajectories using position control. This approach is suitable as long as the accuracy of the robot and the calibration of the workcell is sufficiently precise. In robotic assembly these conditions are usually not fulfilled because of uncertainties, e.g., variability in involved parts and objects not gripped accurately. Using force control is one way to handle these difficulties. Here, two methods of doing force control without a force sensor is evaluated. The methods are based on force estimation from input-output data during closed-loop control with errors and compensations caused by contact forces. The approaches were experimentally verified in a small part assembly task with a kinematically redundant robotic manipulator.