Abstract: The sample data of the rotating center of the alignment platform collected in the process of motion calibration for the bonding equipment has so many interferences,which cause a lot of problems like complexity,low accuracy and robustness in derivation process.In order to solve the problem, figures out a simple solution based on Random Sample Consensus(RANSAC)algorithm.By carrying out iteration screening several times with the solution based on the density distribution model of the sample data,we can derive the results on the basis of the optimum solution.The experimental results show that by using iterative calculation within 10 groups of random sample datum,it costs 697.6 seconds on average,and the calibration accuracy of this method is within 7.2 pixels. Besides,the results of the 7 datasets all meet the precision requirement of the 10 pixels.The results reach the same performance level of the Japan and South Korea’s equipment and absolutely conform to the requirements of the equipment operation.

Key words: bonding equipment, motion calibration, Random Sample Consensus(RANSAC), the density distribution model of the sample data