How might ants use panoramic views for route navigation?

Abstract

SUMMARYStudies of insect navigation have demonstrated that insects possess an interesting and sophisticated repertoire of visual navigation behaviours. Ongoing research seeks to help us understand how these behaviours are controlled in natural complex environments. A necessary complement to behavioural studies is an understanding of the sensory ecology within which an animal behaves. To this end we have analysed ants'-perspective views of a habitat within which desert ant navigation is well studied. Results from our analysis suggest that: parsimonious visual strategies for homing and route guidance are effective over behaviourally useful distances even in cluttered environments; that these strategies can function effectively using only the skyline heights as input; and that the simplicity and efficacy of using stored views as a visual compass makes it a viable and robust mechanism for route guidance.

Insect-Inspired Visual Navigation for Flying Robots

“…To get an understanding of the information that is present in images for homing, we follow the procedures of [10,11,17] and use simple metrics to estimate the range over which a single image can be used either as an attractor type snapshot, to recover a direction towards a goal, or as a visual compass, to recall the heading at which the agent was facing when the goal image was stored.…”

“…To assess the range over which a single image can be us ed for navigation, we quantify the region around a goal image in which the difference between images increases with increasing distance from the reference image. The presence of an increasing IDF is significant as it shows that the information needed for view-based homing is available [10,17]. We therefore define the catchment area (CA) of the IDF as the region where the IDF is generally increasing.…”

“…We therefore next determine what we termed rotational catchmen t areas (RCAs) [17], which indicate the region in which an agent would be able to use a goal image to recall the heading it was facing when the goal images was stored.…”

“…The minimum value in the RIDF defines an orientation of the current image which gives the closest match with the reference image. In the vicinity of the reference image these orientations will be similar to the reference image orientation [10] and so can be used to recall a heading or a movement direction for a forward facing agent [17]. We define the rotational catchment area as the region spreading out from the location of the reference image where the minimum in the RIDF is less than 45° from the true orientation of the reference image (Fig.…”

“…To assess the whether this model can also be used for a flying robot, we thus need to test the model with images gathered from different heights. To do th is we follow the methods of [10,11,17] to first assess the extent over which single images gathered from different heights through two natural environments can provide information for visual homing. We show that in line with Zeil [10], a snapshot stored at one height can successfully be used as either an attractor snapshot or visual compass for an agent travelling at a different height.…”