Pose error analysis method based on a single circular feature

The measurement accuracy of pose parameters based on a single circular feature depends not only on the accuracy of camera calibration and feature extraction but also on the relative pose of the feature and camera—different poses correspond to different error transmission coefficients. To obtain the relationship between measurement errors and pose parameters, we propose an error analysis method based on geometric interpretation. The method characterises measurement error by the sensitivity the imaging feature has to the variation of pose parameters. In addition, the method can be extended to the error analysis work of other coplanar features' pose measurement algorithms. We conducted simulations on measurement errors of pose parameters under different poses, and the results show that the error distribution of pose parameters is in good agreement with the theoretical analysis. Moreover, we propose a method for judging and optimising outliers, and experimental results show the feasibility of this method.