Dynamic Positioning and Navigation
SYNOPSIS OF PROCEEDINGS
The session on GPS dynamic positioning and navigation covered a broad range of applications with diverse requirements for both accuracy and sampling rates of spatial and temporal parameters. A major theme was the use of GPS positioning and navigation for remote sensing applications. Unlike presentations during the session on remote sensing of the atmosphere, however, the focus was not on using GPS receivers as sensing instruments. Instead, the focus was on using GPS to position and navigate moving platforms equipped with optical, gravimetric, or magnetic sensors. Papers were presented during this session on the real-time positioning and navigation of oceangoing acoustic probes, airborne sondes, aircraft, and even satellites. For each of these dynamic applications, there is a subtle, yet important, distinction between positioning and navigation. Navigation involves not just the passive determination of position, but also active steerage and control.
Robin Bell addressed several applications in the realm of aerogeophysics, the goals of which are to acquire systematic, high-resolution data sets, especially for inaccessible areas, such as Antarctica. Measurements include gravity, electromagnetics, topography, and ice penetration. A related presentation by William Krabill and Chreston Martin addressed ice-sheet mapping with airborne lidar, radar, and a scanning-laser topographic mapper. Combining dual-frequency carrier phase GPS receivers and inertial navigation units on board aircraft platforms enables these instruments to produce ice-sheet elevation maps with spatial accuracies in the range of 10 to 20 centimeters.
George Born and Kevin Key discussed using the GPS for satellite altimetry to measurement of changes in global sea level. Specifically, GPS positioning data is used to calibrate and validate satellite altimeter data and to determine corrections for ionospheric path delay. To achieve the required global coverage, the authors suggested that either more dual-frequency GPS ground stations would be needed in ocean areas or that ships of opportunity would have to be equipped with similar receivers. Another alternative mentioned was using ionospheric data collected from space-based systems, such as GPS/MET.
In a related paper, Robert Shutz discussed using GPS for the centimeter-level orbital positioning of LEO satellites. One example mentioned was the Topex/Poseidon altimetry satellite. This satellite carries a dual-frequency GPS receiver that has been used to demonstrate the feasibility of GPS-based orbital positioning as accurate as 2 to 3 centimeters radial and better than 10 centimeters horizontal (along-track and cross-track). In order to achieve this level of accuracy, however, the required differential GPS measurements must be collected using a ground-based receiver network with a stable and highly accurate coordinate system. The network must also utilize all available techniques to eliminate other sources of error. Other examples of LEO satellites that require orbital positioning accuracy better than 1 centimeter in the horizontal plane include satellites used for gravimetric and geodetic research. Thomas Yunck discussed the overall network and data implications of this required accuracy level.1 He pointed out that the FAA's WAAS is expected to have a positive impact on meeting this need.
Kevin Leaman discussed the use of DGPS navigation for physical oceanography. He focused on using a GPS-based acoustic probe, known as Fast Pegasus, to determine absolute, horizontal velocity profiles of ocean currents. For this application, geodetic accuracy, resolution, repeatability, and high sampling rates are all important. Therefore, the surface-drifting buoys used to deploy the Fast Pegasus probes are positioned and tracked using a temporary DGPS network consisting of receivers on the buoys and a reference station on shore. The meter-level accuracy of the buoy tracking enables ocean current velocity profiles to be measured at the centimeter-persecond level of accuracy. An expanded network of GPS ground stations and the availability of real-time corrections would enhance this oceanographic application and enable others.
Dave Carlson, Don Lenschow, and Hal Cole addressed the use of GPS navigation for measuring atmospheric velocity from aircraft and radiosondes. 2 Although standard
No paper corresponding to this presentation is available for publication.
No paper corresponding to this presentation is available for publication.
(nondifferential) GPS receivers can be used to track radiosondes making simple wind measurements, their level of accuracy is not sufficient for measuring boundary-layer vertical velocities from aircraft. This application also requires rugged and compact GPS devices that are low in cost.
In a final paper and videotape presentation, Paul Montgomery demonstrated the success of using differential GPS for the navigation of a small (12-foot) autonomously piloted aircraft. Carrier phase-based differential techniques were used to determine the aircraft's position, velocity, attitude, and angular velocity relative to the ground in real time. Additional data related to air velocity were gathered using on-board sensors. The real-time combination of this information made automated flight from takeoff to landing possible.
WORKING GROUP DISCUSSIONS
GPS was designed with military and civilian dynamic positioning and navigation applications in mind and has been used in this capacity for a number of years. Nevertheless, the first issue addressed by the working group on dynamic positioning and navigation was to define a “dynamic application” in contrast to a “static application.” There was broad agreement that dynamic applications involve vehicles or platforms designed to move on or above the Earth's surface on human time scales. Platforms include satellites, aircraft, trucks, cars, trains, boats, ships, buoys, drifters, dropsondes, and, potentially, autonomous underwater vehicles.
Earth, oceanic, and atmospheric applications using these vehicles or platforms include the measurement and mapping of gravity, magnetics, topography, radiometry, meteorology, and fluid dynamics. Dynamic applications often require real-time measurements of position and attitude to support remote sensors or closed-loop control systems. Real-time augmentation of the basic GPS and high sampling rates are usually required as well.
The great diversity of requirements for this broad range of applications, generated considerable discussion about spatial and temporal accuracy and sampling rates. The spatial resolution requirements discussed were within a range of 1 centimeter to 5 meters for most applications, and temporal resolution requirements were within a range of 0.1 seconds to 5 seconds. However, certain applications have much more stringent requirements. For example, airborne gravimetrics could benefit from spatial resolutions of less than 1 centimeter to derive aircraft accelerations, and sampling rates for some mapping applications could exceed 10 Hz. These requirements are shown in Table 4-1 and Table 4-2.
TABLE 4-1 Spatial Resolution Requirements for Dynamic Applications a
Table 4-2 Temporal Resolution Requirements for Dynamic Applications a
The working group distinguished between the high temporal sampling rates needed at the platform receiver and the sampling rates of GPS observables at reference stations, which are used to determine differential corrections. It was generally agreed that 5-second sampling of correction observables is adequate because factors that influence most known errors do not change very rapidly and because corrections can be interpolated to higher resolution. However, more analysis is needed to determine the sensitivity of final positioning accuracy to the sampling rate of observables used for augmentation. These analyses are important because high-update-rate, real-time corrections could strain the resources of the existing and planned GPS network infrastructure. For example, the real-time positioning and navigation of mobile platforms operating over a wide area requires a wireless broadcast system to disseminate correction data, unlike the batch processing and file transfer methods used in static applications. Because communications bandwidth is an important consideration for wireless broadcasts, efficient data formats must be considered besides the standard receiver-independent exchange (RINEX) format. Minimizing the latency of broadcast corrections is also of considerable importance to real- and near-real-time applications.
For some applications, global GPS network coverage is always desirable and sometimes essential. However, better coverage in critical areas, such as the oceans, may be more important. Incorporating meteorological instrumentation into network reference stations was also discussed. Although meteorological data are not required for most dynamic applications, the data could be used to improve the accuracy of positioning and navigation.
There was general agreement that a second GPS signal dedicated to civilian use would be beneficial. Aside from spectrum allocation, the choice of a center frequency for this new signal should be determined by assessing the trade-offs between separation from the L1 and L2 signals and bandwidth. The working group did not attempt to specify an optimal frequency but generally agreed that a center frequency above both L1 and L2 (1575.42 Mhz and 1227.60 Mhz, respectively) would be desirable. However, because wide bandwidth is important for high resolution and noise reduction, this should take precedence over placement at a higher frequency.