A Low Cost, Hovering, Novel, 2-DOF Micro AUV

Baby AUV is a low cost, hovering, novel 2-DOF micro AUV which will be capable of multi-day environmental monitoring missions, in particular for taking CTD casts at high spatial resolution. The estimated single-unit BOM cost is $500. Apart from CTD, Baby AUV carries GPS and a 3D compass/accelerometer.
     The moving mass actuator controls pitch, and there is a single 1.5 W surge thruster. There are no fins and no sail. Removing these vulnerable parts keeps the cost and drag down. A flexible GPS antenna is carried in the nose.
     With the AUV pitched nose high, hovering is achieved by balancing reverse thrust against reserve buoyancy. The thruster torque reaction causes the AUV to rotate about the vertical and describe a tight orbit, so the AUV will not tend to drift away but remain in the same water parcel. It’s like a helicopter that has lost its tail rotor.
     Without fins or multiple thrusters to provide yaw, Baby AUV controls heading in 2 different and novel ways.
Figure 1. Baby AUV: $500, 3.5kg, 50m,
48 hours, 1m/s, CTD.
Figure 2. Coarse heading sequence shown
(L-R, T-B) covers 15 seconds. The off-centre
thruster is due only to space constraints.
     Coarse heading control (see figure) begins as for hovering, but with greater reverse thrust to corkscrew downwards at a high rotational rate. At the appropriate point, the vehicle pitches level and drives forward on a new heading. The corkscrew can complete in a few seconds and results in a turning circle radius of almost zero. Typically it will take the AUV to within 30 degrees of the desired heading.
     After it gets under way, Baby AUV can trim its heading by appropriately modulating the torque to the single-bladed propeller, to develop a net side force. Put another way, the asymmetry induces in the AUV a left/right wiggle, like a fish but less pronounced. By raising the torque when the blade moves from 9 to 3 o’clock and reducing it from 3 to 9 o’clock, the stern of the AUV wiggles left more than right, resulting in a slow left turn, for example.
     This is like prop walk in power boats, except here it is the propeller and not the flow condition that is asymmetric, but the effect is the same.
     While it may appear odd, single bladed propellers are occasionally used in aviation. It is also found in nature as the bacterial flagellum (whip-like tail), and in fact tumbling motility in bacteria is strikingly similar to coarse heading control described above.
     Apart from taking a minimalist approach to actuation, cost is reduced by restricting machining to mostly the end bulkheads. The hull itself is a standard aluminium extrusion. O-rings are eschewed because they require tight tolerances and a controlled surface finish. Instead, the bulkheads carry a neoprene expansion seal which permits assembly and disassembly without special tools.
     Endurance is helped by the lack of hull protrusions into the flow (no fins, no sail), from a high slenderness ratio of 6, and from the efficiency gains of the large and slow propeller, which gains are predicted from momentum theory. Not coincidentally, a large and slow propeller is also needed to generate enough torque for effective coarse heading control. This contrasts strongly against the usual AUV practice of running small propellers at high speed and regarding thruster torque reaction as an embarrassment.
     There are other volume–or area-covering missions suitable for Baby AUV that do not require high path or position accuracy. Apart from electrical conductivity, the sensors for pH, dissolved oxygen, ORP, and PAR, are low in power and fairly economical. Probably 1 or 2 such sensors can be integrated per AUV in addition to depth.
Harold Tay
     Algal bloom monitoring would appear to be a perfect use case, but the fact that a chlorophyll sensor costs several times the rest of the AUV gives one pause.
     Acoustic monitoring is attractive for reef monitoring, harbour protection, for anti-poaching, etc. and can be quite economical, but work will be needed to reduce the power requirement.
     More information on Baby AUV may be found at http://arl.nus.edu.sg/twiki6/bin/view/ARL/HaroldTay.
     The author graduated from the University of Texas at Austin in mechanical engineering. He has worked as mechanical, software, and systems engineer at various times. Now he also works in electronics in his capacity as senior engineer at Acoustic Research Lab of the Tropical Marine Science Institute, National University of Singapore. He is a member of IEEE & OES and has been involved with the Singapore AUV Challenge for the past 3 years and as its chair for the 2016 competition. He has also contributed to a beacon article earlier. See details at http://www.oceanicengineering.org/userfiles/files/June-2016-OES-Beacon.pdf. Harold also served on the executive committee of the IEEE OES Singapore Chapter in 2016.


If you would like to contact the IEEE Webmaster
© Copyright 2016, IEEE. Terms & Conditions. Privacy & Security

return to contents