Learning Human-Inspired Force Strategies for Robotic Assembly
This addresses the challenge of programming force-sensitive robotic assembly tasks in manufacturing, offering a method to learn strategies offline without real hardware, though it is incremental in combining existing techniques.
The paper tackled the problem of learning reactive force strategies for robotic assembly from human demonstrations in simulation, achieving a UR10e robot that completed a plastic assembly with clearances under 100 micrometers using strategies learned solely from simulation.
The programming of robotic assembly tasks is a key component in manufacturing and automation. Force-sensitive assembly, however, often requires reactive strategies to handle slight changes in positioning and unforeseen part jamming. Learning such strategies from human performance is a promising approach, but faces two common challenges: the handling of low part clearances which is difficult to capture from demonstrations and learning intuitive strategies offline without access to the real hardware. We address these two challenges by learning probabilistic force strategies from data that are easily acquired offline in a robot-less simulation from human demonstrations with a joystick. We combine a Long Short Term Memory (LSTM) and a Mixture Density Network (MDN) to model human-inspired behavior in such a way that the learned strategies transfer easily onto real hardware. The experiments show a UR10e robot that completes a plastic assembly with clearances of less than 100 micrometers whose strategies were solely demonstrated in simulation.