A novel approach for robot calibration based on measurement sub-regions with comparative validation

This paper presents a methodology for industrial robot calibration using measurement points collected in several sub-regions within the robot workspace separately and extends the effectiveness of this proposal to calibration in a large workspace where the points are all obtained from a fixed location, with wide variation of the robot joint positions. The identification process will estimate the error parameters associated with each of these sub-regions, obtaining a corrected kinematic model related to each of them. A new corrected kinematic model was statistically calculated from the error parameter vector of each sub-region that will meet the accuracy requirements of the robot calibrated over a large workspace. The experimental procedures were performed with a Laser-Tracker measurement system and a Fanuc 200i industrial robot, validating the effectiveness of the sub-region approach. The proposed methodology was experimentally demonstrated to have applicability in industry due to its results, turning feasible the use of a simpler and cheaper measurement system that can be used in congested environments or with limited space, ensuring accuracy equivalent to calibration with wide robot movements in large work volumes and with high-precision sensors for measuring at long distances. The accuracy obtained with calibration in sub-regions was compared with calibration using all the points in a large workspace, reaching 0.70 mm and 0.84 mm, respectively. The results show that calibration based on sub-regions is a suitable alternative for the accuracy requirements of specific applications, such as in the automotive and aerospace industries, or where measurement in small volumes may be more viable.