Parameter Uncertainty

The parameter uncertainty is estimated by a variance for each of the six transformation parameters (3 for the translation offset and 3 for rotation). The parameter uncertainty is estimated by adapting the Jackknife resampling method. It should be noted that the results obtaind are only an approximation of the actual variance of the parameters, which are also only a statistical measure of the actual error in the transformation.

JACKKNIFE RESAMPLING^

Jackknife resampling was originally suggested by Quenouille [1]. The main idea of the method is obtaining Jackknife samples. These are simply obtained by removing one data reading from the entire sample. Consequently, the number of samples will mount to be equal to the number of data readings. For each sample, the parameters are calculated and the entire set of parameters is used to calculated the standard error of the parameters (For details on the implementation of the Jackknife resampling method, refer to [2]).

References:

1. M. H. Quenouille. Approximate Tests of Correlation in Time-Series. Journal of the Royal Statistical Society. Series B (Methodological), 1949. 11(1): p. 68-84.

2. T. Peynot and A. Kassir. Laser-camera data discrepancies and reliable perception in outdoor robotics. In Proceesdings of the IEEE/RSJ International Conference on Intelligent Robots and Systems, 2010.