End-Effector Pose Correction for Versatile Large-Scale Multi-Robotic Systems

Abstract

In this paper, we present a fully-integrated end-effector positioning system for large-scale multi-robotic setups used for non-repetitive manufacturing tasks. The system provides static and dynamic correction of the end-effector pose using an external pose tracking system. It consists of multiple modules, which extend the capabilities of the conventional robot setup and fundamentally improve its usability and efficiency, since the user can easily set up, execute, and monitor the manufacturing tasks in a universal reference frame. To increase the performance of closed-loop control of the end-effector pose, a sensor fusion algorithm is implemented for fusing the data of the tracking system iGPS with an IMU. Experiments are carried out, which show a reduction of the average error down to 0.10 mm for the static correction, a significant increase of the measurement quality of the tracking system with the sensor fusion, and a path error below 0.5 mm for the dynamic correction. The presented correction system enables new applications in digital building construction, which require high accuracy target poses spread throughout a large workspace of cooperating robots.