Session: 02-09-01: Session #1: Variation Simulation and Design for Assembly Description

Paper Number: 94705

94705 - A Theoretic Error Modeling Method for the Robotic Optical Sensing System With a 3D Laser 

Robotic optical inspection systems have gradually attracted widespread attention due to their non-contact, inspection flexibility, and high adaptability to manufacturing environments. However, compared with the high-precision coordinate measuring machine(CMM), the higher error of the optical inspection system is the main problem restricting its application. The existing studies mainly focused on the experiment-based error transfer model, or the theoretic error modeling with some of the error sources of the robotic optical sensing system. In this paper, a systematic analysis of the error sources in the optical sensing system is performed. For example, the robot poses error, hand-eye error and the sensor error, etc.  Based on the above error source analysis, a theoretic error modeling method for the robotic optical sensing system with a 3D laser scanner is proposed in this paper. 
The modeling process of the robot inspection system can be roughly divided into the following three steps. Firstly, we analyze the imaging principle of the line laser sensor and determine the two factors that affect the sensor error: incident angle and measurement height. Furthermore, the error model between the incident angle, the measurement height and the sensor measurement value was established.
Secondly, on the premise that the initial hand-eye matrix of the robot is known, an error model of the robot measurement system including hand-eye error, robot error and sensor measurement error is established, and the upper bounds of the deviation of different measurement points is determined based on the tolerance requirements of the measured points. Further, based on the constraint of the common spherical center and the maximum deviation of the measurement points, the optimal robot link parameter error and optimized hand-eye parameters are solved through an iterative algorithm.
Finally, based on the optimized geometric parameters and the tolerance requirements of different measurement points, the measurement pose and measurement height of the robot under different measurement points are optimized to ensure that the error of different measurement points meets the requirements. 
In order to verify the effectiveness of the proposed method, we use the Gocator2450 sensor mounted on the UR5 robot to inspect different measurement points. Based on the proposed error transfer model of the measurement system, the measurement pose and measurement height are optimized, and CMM is used to verify the accuracy.
Based on the error model of optical inspection system proposed in this paper, it can not only improve the inspection accuracy of a single measurement point but also provide guidance for robot coverage path planning to inspection pose and height, so as to improve the inspection efficiency of the robot and ensure the measurement accuracy at the same time. In conclusion, the proposed error model can provide a theoretical basis for a wide range of applications in optical detection systems.
 

Presenting Author: Wenzheng Zhao University of Shanghai for Science and Technology

Presenting Author Biography: Wenzheng Zhao obtained the M.S. degree in Mechanical engineering, University of Shanghai for Science and Technology, Shanghai, China, in 2021. He is currently working toward the Ph.D. degree in Mechanical Engineering at University of Shanghai for Science and Technology, Shanghai, China. His research interests include assembly quality control and robotics optical inspection of sheet metal assemblies. He has published some journal papers in IEEE-ASME Transactions on Mechatronics, Journal of Manufacturing Systems, Journal of Intelligent Manufacturing etc., 3 China invention patents, and got the “Excellent Ph.D student in Intelligent Manufacturing ” for his programming capability in the development of the software tool “Process planning of the optical robotic sensing in the assembly process” of Shanghai General Motors.

Authors:

Wenzheng Zhao University of Shanghai for Science and Technology
Yanzheng Li University of Shanghai for Science and Technology
Yinhua Liu University of Shanghai for Science and Technology