Mimicking the elegance of insect behavior: a case study in visual navigation
Biomimetic control systems in engineering and autonomous robotics have shown lots of promise but depend on a full understanding of the target biological system. For a complete understanding of natural behavior, it is essential to understand the constraints placed on behaviour-generating mechanisms by morphology, sensory systems and the information available in the natural habitat. That is, one must understand the animal as a situated and embodied system. Only then can one reap the full benefits of biomimickry. Our work has been driven by trying to understand these constraints in the context of visual homing in ants. In this talk, I will present a summary of our work and show that by considering the ant’s natural behavior sensory constraints and natural visual ecology, we have developed a simple system capable of navigating ant-like routes through complex habitats with very limited computational and sensory resources.
We have shown through behavioural studies and modelling that the 1D skyline profile between terrestrial objects and the sky is sufficient for ants to set a direction and contains enough information to guide a route. During these, and other, experiments we observed that when ants are choosing a direction they scan the world by pausing and rotating on the spot. The scanning is saccadic in nature and ants' fixations are drawn by familiar portions of the panorama. Following these results, we propose a novel parsimonious algorithm for visually guided route following in which agents scan the environment and move in the direction that appears most familiar. We return to behaviour by showing how the structure of the learning walk of ants can be understood in the context of this model.