Route Learning Through Classification

A simple approach to route following is to scan the environment and move in the direction that appears most familiar . In this paper we investigate whether an approach such as this could p rovide a model of visually guided route learning in ants. As a proxy f or familiarity we use the learning algorithm Adaboost [6] with simple Haar-like features to classify views as either part of a lear n d route or not. We show the feasibility of our approach as a model of an t-like route acquisition by learning a non-trivial route through a re l-world environment using a large gantry robot equipped with a panor amic camera.

[1]  R. Rosengren,et al.  Ortstreue in foraging ants of theFormica rufa group — Hierarchy of orienting cues and long-term memory , 1986, Insectes Sociaux.

[2]  Lincoln Smith,et al.  Linked Local Visual Navigation and Robustness to Motor Noise and Route Displacement , 2008, SAB.

[3]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

[4]  T. Collett,et al.  Local and global vectors in desert ant navigation , 1998, Nature.

[5]  R. Wehner,et al.  Visual navigation in insects: coupling of egocentric and geocentric information , 1996, The Journal of experimental biology.

[6]  R. Wehner The architecture of the desert ant's navigational toolkit (Hymenoptera: Formicidae) , 2009 .

[7]  R. Wehner,et al.  Navigation in wood ants Formica japonica: context dependent use of landmarks , 2004, Journal of Experimental Biology.

[8]  Frédéric Labrosse,et al.  The visual compass: Performance and limitations of an appearance‐based method , 2006, J. Field Robotics.

[9]  T. Collett,et al.  Snapshot Memories and Landmark Guidance in Wood Ants , 2003, Current Biology.

[10]  R. Wehner,et al.  Traveling in clutter: Navigation in the Central Australian desert ant Melophorus bagoti , 2009, Behavioural Processes.

[11]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1995, EuroCOLT.

[12]  Lincoln Smith,et al.  Linked Local Navigation for Visual Route Guidance , 2007, Adapt. Behav..

[13]  T. Collett,et al.  Visual landmarks and route following in desert ants , 1992, Journal of Comparative Physiology A.

[14]  T. S. Collett,et al.  Landmark learning in bees , 1983, Journal of comparative physiology.

[15]  Jochen Zeil,et al.  Catchment areas of panoramic snapshots in outdoor scenes. , 2003, Journal of the Optical Society of America. A, Optics, image science, and vision.

[16]  R. Wehner,et al.  Visual spatial memory in desert ants,Cataglyphis bicolor (Hymenoptera: Formicidae) , 1979, Experientia.

[17]  Rüdiger Wehner,et al.  Idiosyncratic route-based memories in desert ants, Melophorus bagoti: How do they interact with path-integration vectors? , 2005, Neurobiology of Learning and Memory.