Lossy compression and real-time geovisualization for ultra-low bandwidth telemetry from untethered underwater vehicles

Abstract

Oceanographic applications of robotics are as varied as the underseaenvironment itself. As underwater robotics moves toward the studyof dynamic processes with multiple vehicles, there is an increasingneed to distill large volumes of data from underwater vehicles anddeliver it quickly to human operators. While tethered robots areable to communicate data to surface observers instantly, communicatingdiscoveries is more difficult for untethered vehicles. The oceanimposes severe limitations on wireless communications; light is quicklyabsorbed by seawater, and tradeoffs between frequency, bitrate andenvironmental effects result in data rates for acoustic modems thatare routinely as low as tens of bits per second. These data ratesusually limit telemetry to state and health information, to the exclusionof mission-specific science data.In this thesis, I present a system designed for communicating andpresenting science telemetry from untethered underwater vehiclesto surface observers. The system’s goals are threefold: to aid humanoperators in understanding oceanographic processes, to enable humanoperators to play a role in adaptively responding to mission-specificdata, and to accelerate mission planning from one vehicle dive tothe next. The system uses standard lossy compression techniques tolower required data rates to those supported by commercially availableacoustic modems (O(10) – O(100) bits per second).As part of the system, a method for compressing time-series sciencedata based upon the Discrete Wavelet Transform (DWT) is explained,a number of low-bitrate image compression techniques are compared,and a novel user interface for reviewing transmitted telemetry ispresented. Each component is motivated by science data from a varietyof actual Autonomous Underwater Vehicle (AUV) missions performedin the last year.