Photogrammetry based error analysis of indoor mobile robot localization

The presented work is a low cost “off-line” method that is able to evaluate the precision of mobile robot localization, which is a difficult task. A reference path is yielded by a drawing pen attached to indoor mobile robot, which also adds timestamps to the trace on the floor. The method is based on comparing the reference path obtained using photogrammetry, with the results of several localization algorithms on board of mobile robots. At the moment, different techniques exist that deal with that problem, but these are either expensive or too rigid.

[1]  Zhengyou Zhang,et al.  Flexible camera calibration by viewing a plane from unknown orientations , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[2]  Alonzo Kelly.  Mobile Robot Localization from Large-Scale Appearance Mosaics , 2000, Int. J. Robotics Res..

[3]  Janne Heikkilä,et al.  A four-step camera calibration procedure with implicit image correction , 1997, Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Carme Torras,et al.  Using Laser and Vision to Locate a Robot in an Industrial Environment: A Practical Experience , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[5]  Hugh F. Durrant-Whyte,et al.  Inertial navigation systems for mobile robots , 1995, IEEE Trans. Robotics Autom..