Prev-Talk | Next-Talk | All-Talks | Talks-Sorted
Talk-18: Navigation and autonomy for low-cost autonomous underwater vehicles using MOOS-IvP
Erin Fischell, Woods Hole Oceanographic Institution
Nick Rypkema, MIT / WHOI Joint Program
A new class of low-cost AUVs, including General Dynamics SandShark vehicle, make multi-vehicle sensing more realistic from a cost perspective. With each vehicle in the tens of thousands rather than hundreds of thousands or millions of dollars, riskier autonomy modes and adaptive networks or formations of vehicles seems a nearer reality. However, there is an element of "you get what you pay for" with these AUVs: they lack sensors used on more conventional vehicles to improve navigation and communication, such as Doppler velocity logs (DVLs), Inertial Navigation Sensors (INS), or acoustic modems. We are developing an AUV operational paradigm in MOOS-IvP geared specifically for vehicles lacking high-fidelity navigation or communication hardware: instead of operating in an absolute coordinate system, we re-define the autonomy and sampling problem in relative coordinates. Vehicle behaviors are defined relative to an acoustic beacon in the range/bearing space that comes from array processing of acoustic data. Each AUV carries simple acoustic receiver payloads with no active acoustic elements to limit power draw and restrict the noise being put into the environment so that large numbers of vehicles can operate simultaneously. Communications are simplified and limited, so that the waveform from the beacon commands formation-wide behaviors. Simulation and experimental results on a SandShark AUV are presented.
Categories:
- AUVs
- Multi-Robot Operations