Meeting Abstract
The lateral line system of fishes is one of the most elaborate and ancient sensing architectures found in nature. It consists of mechanoreceptors distributed around the body which provide information on local flows and pressure gradients. Neurophysiological and theoretical studies have revealed important insights into the function and capabilities of individual receptors, while organismal studies have taken advantage of pharmacological treatments to ablate the lateral line system in order to interpret its function during critical behaviours such as rheotaxis, predator evasion, prey tracking and obstacle avoidance. What remains missing is an integrated view on how hydrodynamic information is organized around swimming fishes. During swimming, extracting meaningful information from local flows is challenging because fish must be able to distinguish between an external stimulus and a stimulus generated through its own movements. Here, we leverage a combination of robotics, modelling, and biological experiments to investigate what information is available to the distributed architecture of the lateral line; specifically, how head morphology and swimming movements affect the organization of this information. Our initial efforts focus on the 3-dimensional fluid-structure interactions around the head of rainbow trout (Oncorhynchus mykiss), which has an intriguing canal pattern compared to the body. Our results show that 1) there exist sub-regions within the cranial lateral line system that are uniquely sensitive to different hydrodynamic stimuli, 2) the shape and surface topography of the fish head act like a hydrodynamic antenna to enhance lateral line sensing, and 3) sensing volume and frequency response of the head changes dynamically with the swimming patterns of the fish. Our findings promise to provide novel design principles for distributed sensing systems in underwater robotics.
