Navigation for autonomous underwater vehicles (AUVs) could become more accurate with the development of a $100 sensory device that mimics the ‘feelers’ found on a blind deep water fish.
Engineers at Nanyang Technical University in Singapore have prototyped the sensors, which use a combination of computer vision and water pressure technology to create a 3D image of their surroundings, and are testing them on AUVs developed by partners Singapore-MIT Alliance for Research and Technology (SMART) for environmental monitoring.
The low-energy MEMS (microelectromechanical systems) devices could make the often bulky, inaccurate and heavily energy-dependent cameras and sonars usually integrated onto AUVs redundant, instead packing all the navigational tools into an array of 1.8mm-square sensors.
“Other methods like underwater lights and cameras, acoustic navigation, and sonars do work, but they are very expensive and require very high levels of power that drain the batteries,” says Professor Michael Triantafyllou of Massachusetts Institute of Technology (MIT). “The new sensors are much cheaper and only require small amounts of power. Also, sensors like sonar are loud and invasive and they may harm aquatic animals that also use sound waves to navigate.”
A team led by Associate Professor Miao Jianmin created arrays of tiny sensory pillars coated with a hydrogel to mimic the feelers on each side of the blind cave fish that allow it to move around the sea-bed without collisions. A piezo-electric sensor generates a voltage when water flows past the feelers, while a low-powered biomimetic sensor can detect objects even in still water.
Prof Miao explained: “this array of micro-sensors will allow AUVs to locate, identify, and classify obstacles and objects in water through water pressure and also to optimise its movement in water by sensing the water flow.”
External link: http://eandt.theiet.org/news/2013/jan/fish-sensor.cfm#