The purpose of this work is to implement and study the dynamics of a modified version of a biomorphic visual navigation system. 

Methods. The paper uses the RatSLAM simultaneous navigation and mapping system. The RatSLAM system is a biomorphic model of visual navigation in the rodent hippocampus. In this study, we investigate a modified version of the RatSLAM system in which visual landmarks are processed using a two-threshold algorithm.

Results. This article presents a modified version of the visual navigation system. Using a two-threshold algorithm for determining visual landmarks allows for a reduction in the size of the resulting map without loss of accuracy. Using the constructed system, location estimates and clustering metrics for visual landmarks were obtained on publicly available datasets.

Conclusion. The constructed visual navigation system provides an estimate of the location of an object (video camera) in space that is in good agreement with the true location data. Using a two-threshold algorithm, the map size can be reduced without increasing map accuracy.