Whisker Sensor Geometry

Sensing underwater presents additional challenges compared to land due to the constantly changing environment, low visibility, and turbidity constraints. Hence, camera and laser based technologies that work on land do not work well in the water due to high absorption rates. Furthermore, many technologies for underwater sensing are expensive and have high power requirements, which limit the additional functionalities of the AUV. Thus, navigating underwater is a hassle.Our goal is to take inspiration from harbor seal whiskers to develop a low cost, low power sensor that can detect changes in underwater flow to aid in navigation. Harbor seals have unique wavy whiskers with asymmetric oscillations along their profiles, an elliptical cross section, and a material stiffness similar to uncooked pasta [1] [2]. Heather Beem studied this geometry and determined that the waviness allows seals to break up the flow as they swim, but they vibrate when interacting with small vortices (e.g. from prey) [1] [3]. Hence, the whiskers are steady until interacting with an environmental change, which can be used for navigation.Based on Heather Beem’s work, Professor Pablo Valdivia y Alvarado of the Singapore University of Technology and Design developed a piezoresistive sensor similar to seal whiskers. He sent a collection of sensors to MIT Sea Grant Laboratory for us to experiment with, use on AUVs, and get inspiration from. In our paper, Whisker Sensor Calibration and Replication, we discuss our calibration experiment of Alvarado’s sensors where we determined that they should be used to detect relative changes in voltage to indicate a change in flow [2]. Furthermore, we present our current work related to base manufacturing for making our own sensors based on what we learned in the experiment [2].For the whisker element of our sensor, we are experimenting with different shapes to achieve the same results as the seals: steady as the AUV moves in the water but vibrating when interacting with another vortex. To determine the proper geometry, we started by CADing different options inspired by the work done by other scientists, Heather Beem’s tow tank experiment whisker (based off of in-lab measurements), descriptions by Patricia Leonard (Supervisor of Pinnipeds at the New England Aquarium), and physical observations of seals at the New England Aquarium [2]. Each profile geometry used multiple CAD models based on which technique is used to connect the profiles, which was done to account for possible geometric artifacts in fabrication.Next, we resin printed the whisker geometries and conducted a hand test in the tow tank. This informal test included running the whiskers through the water in different orientations (horizontal, vertical, slanted angle of attacks) and with different profiles facing the flow. The vibrations in the whisker can be seen in video recordings and felt in the hand holding onto the whiskers. Based on these tests, we determined that from the perspective of sensing the vibrations in one’s hand, the technique used in CAD, with possible associated geometric artifacts, does not have an effect on whisker performance. Overall, each whisker had a preferred orientation where it performed well. This indicates that using different whisker geometries and arranging them around a AUV could lead to directionality. In addition, observation of seals exhibiting hunting and tracking behaviors have inspired us to explore the effect of the angle of attack on sensor performance as we noticed that the seals keep their whiskers in certain positions with differing AOA for eating and swimming, and we observed differing qualitative hydrodynamic effects in the simple hand test. Using our results, we plan to scale up the different geometries and conduct flow tests in the tow tank.