利用低成本GNSS/IMU浮標監測海洋訊號

Monitoring Ocean Signals Using Low-Cost GNSS/IMU Buoys

海表面高度變化包含不同頻率訊號，例如海嘯、氣象海嘯、海潮、風暴潮以及洋流等，這些變化直接或間接地影響人類生命與財產安全，因此準確監測海水面高成為一個重要的研究課題。現今海水面高度變化都以潮位站、衛星測高、傳統加速度浮標來進行觀測，然而潮位站資料含有地殼變動訊號，衛星測高在沿岸區域觀測精度較低，傳統加速度浮標造價昂貴、體積龐大與受到低頻雜訊的影響，且三者皆無法測得所有頻率之海水面訊號。本研究引入高頻取樣全球導航衛星系統（Global Navigation Satellite System, GNSS）浮標觀測量來觀測海水面變化，另藉由慣性量測單元（Inertial Measurement Unit, IMU）接收之加速度及角速度等資料，提供GNSS訊號遮蔽時資料與浮標傾斜之改正。最後整合GNSS相對定位或精密單點定位成果與IMU觀測量求得水面變化，並將成果與現地波高計、潮位站以及港外波浪觀測資料進行比較。本研究於兩處進行實驗測試，首先於成功大學水工試驗中型斷面水槽內，利用造波機產生固定頻率之規則波，並將GNSS、IMU以及智慧型手機IMU觀測水面變化與波高計資料相互比較，成果顯示觀測之規則波振幅與頻率結果一致。另一實驗地點為台南安平港，GNSS相對定位與安平潮位站資料差值標準差（Standard Deviation, SD）可達1公分等級，精密單點定位解之SD最佳可達3公分左右，而GNSS和IMU整合解精度並無增進。分別利用GNSS、IMU以及智慧型手機IMU觀測水面結果進行能量頻譜分析與有效波高計算，成果顯示除智慧型手機外，GNSS與IMU可有效偵測實際的波浪頻率，但三者皆無法偵測實際波高振幅變化。因實驗地點位處於港內，浮標傾斜較小，且GNSS天線盤離水面僅20公分，故浮標傾斜角改正對觀測結果提升有限。本研究自組的GNSS/IMU浮標，相對於市面銷售之整合儀器價格相對低廉，量測精度也符合海水面觀測所需。 Sea surface heights contain ocean signals with different frequencies, including tsunami, meteotsunami, ocean tides, storm surges and currents, which have direct and indirect impacts on the economy and life of the coastal residents. Therefore, accurately monitoring sea surface height is an important and necessary task. Nowadays, tide gauges, satellite altimetry and the traditional accelerometer buoys are commonly used for monitoring ocean surfaces; however, tide gauge measurements contain crustal deformation, altimetric measurements are inaccurate near the coastal regions, and the accelerometer buoys are expensive, bulky and the measurements are contaminated by low-frequency noises. All of them are not capable of sensing ocean signals with all frequencies. In this study, we built the Global Navigation Satellite System (GNSS)/Inertial Measurement Unit (IMU) buoys to measure sea surface heights. GNSS observations are used for buoy positioning and a small, low-cost and self-assembly autonomous IMU, independently collecting continuous acceleration and angular velocity data, could provide positions when GNSS signals are blocked and tilt corrections of the moving buoys. We integrated the Relative Positioning (RP) or Precise Point Positioning (PPP) solutions with IMU data, and then evaluate the performance by comparing with in situ gauges or sea wave buoy observations. In the study, the experiments were performed in two places. One was conducted in the tank of Hydraulics Laboratory of NCKU, Tainan, and the results show that GNSS and IMU both can detect the frequencies and amplitudes of the simulated regular wave heights, also the observed heights agree with those from the water gauge. The other was conducted in the Anping Harbor, Tainan. The Standard Deviation (SD) differences of DGNSS and PPP solutions can reach 1 cm and 3 cm, respectively, compared with Anping tide gauge. And the GNSS/IMU integrated solutions have the same results. In addition, the GNSS, IMU, and smartphone IMU both have an ability to observe the wave frequencies, but not the amplitudes compared with the wave gauge. The tilt correction of the buoy does not significantly improve the accuracy of the observed heights since the field survey was in the harbor and the height of the antenna height is short at 20 cm only.

Sea surface heights contain ocean signals with different frequencies, including tsunami, meteotsunami, ocean tides, storm surges and currents, which have direct and indirect impacts on the economy and life of the coastal residents. Therefore, accurately monitoring sea surface height is an important and necessary task. Nowadays, tide gauges, satellite altimetry and the traditional accelerometer buoys are commonly used for monitoring ocean surfaces; however, tide gauge measurements contain crustal deformation, altimetric measurements are inaccurate near the coastal regions, and the accelerometer buoys are expensive, bulky and the measurements are contaminated by low-frequency noises. All of them are not capable of sensing ocean signals with all frequencies. In this study, we built the Global Navigation Satellite System (GNSS)/Inertial Measurement Unit (IMU) buoys to measure sea surface heights. GNSS observations are used for buoy positioning and a small, low-cost and self-assembly autonomous IMU, independently collecting continuous acceleration and angular velocity data, could provide positions when GNSS signals are blocked and tilt corrections of the moving buoys. We integrated the Relative Positioning (RP) or Precise Point Positioning (PPP) solutions with IMU data, and then evaluate the performance by comparing with in situ gauges or sea wave buoy observations. In the study, the experiments were performed in two places. One was conducted in the tank of Hydraulics Laboratory of NCKU, Tainan, and the results show that GNSS and IMU both can detect the frequencies and amplitudes of the simulated regular wave heights, also the observed heights agree with those from the water gauge. The other was conducted in the Anping Harbor, Tainan. The Standard Deviation (SD) differences of DGNSS and PPP solutions can reach 1 cm and 3 cm, respectively, compared with Anping tide gauge. And the GNSS/IMU integrated solutions have the same results. In addition, the GNSS, IMU, and smartphone IMU both have an ability to observe the wave frequencies, but not the amplitudes compared with the wave gauge. The tilt correction of the buoy does not significantly improve the accuracy of the observed heights since the field survey was in the harbor and the height of the antenna height is short at 20 cm only.