Three 2D-warping schemes for visual robot navigation

• Published in: Autonomous robots Vol. 29; no. 3-4; pp. 253 - 291
• Main Authors: Möller, Ralf, Krzykawski, Martin, Gerstmayr, Lorenz
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01-11-2010; Springer Nature B.V
• Subjects: Approximation; Artificial Intelligence; Autonomous; Computer Imaging; Control; Cybernetics; Engineering; Error analysis; Homing; Mathematical analysis; Mechatronics; Navigation; Pattern Recognition and Graphics; Production methods; Robotics; Robotics and Automation; Robots; System effectiveness; Vectors (mathematics); Vision; Warpage; Warping; Visual homing; Image warping; Navigation
• ISSN: 0929-5593; 1573-7527
• DOI: 10.1007/s10514-010-9195-y

Warping (Franz et al., Biological Cybernetics 79(3), 191–202,
1998b
) and 2D-warping (Möller, Robotics and Autonomous Systems 57(1), 87–101,
2009
) are effective visual homing methods which can be applied for navigation in topological maps. This paper presents several improvements of 2D-warping and introduces two novel “free” warping methods in the same framework. The free warping methods partially lift the assumption of the original warping method that all landmarks have the same distance from the goal location. Experiments on image databases confirm the effect of the improvements of 2D-warping and show that the two free warping methods produce more precise home vectors and approximately the same proportion of erroneous home vectors. In addition, two novel and easier-to-interpret performance measures for the angular error are introduced.