Submitted to: Applied Engineering in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: April 26, 2005
Publication Date: September 1, 2005
Citation: Smith, L.A., Thomson, S.J. 2005. Gps position latency deetermination and ground speed calibration for the satloc airstar m3. Applied Engineering in Agriculture. 21(5): 769-776 Interpretive Summary: Application systems are being developed that are capable of varying the rate of aerially applied pesticides in response to requirements stored in "prescription files" stored in a control computer. This technology makes use of Global Position System (GPS) coordinates to determine the current spatial position and ground speed of the aircraft and then accesses the "prescription" to find the application rate required for that position. The speed and application rate information is then communicated to an automatic flow control system that adjusts the flowrate of pesticide delivered from the aircraft to the required application rate. A finite period of time is required for the GPS coordinate information to be processed so that position coordinates are available to the computer; therefore, the aircraft (traveling at a nominal speed greater than 200 km/h) no longer occupies the position determined from the information received by the GPS. A position estimate, based on prior positions and ground speed, is sometimes used to better synchronize the position coordinates with the physical position of the aircraft. This study was conducted to measure the difference between physical position of the aircraft and the GPS coordinates available when it occupied that position. A reference point on the ground was established by determining its GPS position coordinates and setting up a rectangular mirror to reflect sunlight in a vertical beam. This light was received from a second mirror angled to the sun and located to the side of the flight path such that it did not interfere with the aircraft operation. An electronic circuit was designed to synchronize the aircraft position with the reference point position. The circuit was installed on the aircraft such that the precise time of day and position of the aircraft was recorded each time it flew over the light beam. These data could then be used to evaluate the difference between the reference point position and the aircraft position when it crossed the reference point. This difference is referred to as position latency. The GPS position coordinates were also used to compute ground speed of the aircraft. Ground speed was calibrated by using two reference points established a known distance apart. Time and ground speed data recorded during the time interval between flying over the first and second reference point were used to compute the average GPS ground speed and the average ground speed (distance/elapsed time). Results indicated that latency magnitudes were less than nine meters, but that the direction of travel had an effect on the type of latency. When traveling in a north or south direction, the GPS position of the aircraft was in front of the actual aircraft position, but when traveling in an east or west direction the GPS position of the aircraft was behind its actual position. Multiple runs in each direction indicated that the measurements were very consistent with expected differences being less than 0.7 m. Ground speed measurements were found to be very accurate and consistent. Percent error for eight runs with a wide range of ground speed magnitudes varied from 0.01% to 0.11%.
Technical Abstract: A SATLOC AirStar M3 swathing system installed on an Air Tractor 402B aircraft was evaluated to determine the GPS (Global Positioning System) position latency associated with known ground coordinates and to determine the accuracy of GPS ground speed. A photo-detector circuit was used to synchronize aircraft position with reference points on the ground from which vertical beams of sunlight were reflected upward. When the aircraft flew over the light beam, the circuit responded by closing a solid-state relay to which the pressure switch cable from the SATLOC system was attached. This action caused an extra record to be inserted into the SATLOC log file with precise time and position information associated with the event. Latency was determined by comparing the aircraft GPS position with the reference point GPS position GPS ground speed was calibrated by using two reference points to close and open the relay, respectively. The difference between the time associated with the relay opening and closing could be used with the measured distance between the points to compute average ground speed. Results indicated that position latency was positive for N-S flight directions and negative for E-W flight directions. The magnitude of these latencies was relatively small (less than 9 m) considering that the aircraft was traveling at a nominal speed of 58 m/s. GPS ground speed calibration results revealed errors ranging from 0.01% to 0.11% for a series of eight runs with speeds varying from 176 to 238 km/h.