How can I reduce the cost of the motion sensor for lake and river work

Motion sensors are very expensive; can I use a cheap one on lakes and rivers, where there isn't much wave action?

When you are surveying on the sea, with the waves moving the boat around, a good motion sensor is needed to measure the motion of the boat. If the motion measurement is not accurate, then you see "motion artefacts" in the measurements of the depths, where the depth map has waves in it that match the movement of the boat.
But if you are on rivers and lakes, there is much less movement of the boat. There is always some motion, and a lake on a windy day can have quite big waves. As people move around in the boat, or when the boat turns, it will roll from side to side enough to make a big difference to the depth measurements. So, you do need some motion compensation.
Bathyswath gives depth measurements over a wide range; even in 10m water depth, you can easily get ranges of 100m. You can work out the accuracy that you need using trigonometry; suppose you can accept a 10cm error: the angle accuracy you need is [arctan(0.01/100)=] 0.05 degrees. That is why we recommend sensors with an accuracy of better than 0.05 degrees.
At sea, heave motion (up and down) is also an important error factor, and needs an accurate sensor. But on lakes and rivers, there is almost no heave if the weather is good, so we don't have to worry about that too much.
Any "static errors", errors in the motion measurement that never change, will be compensated for in the patch test process, so we don't have to worry about those.
The difference between cheap motion sensors and expensive ones isn't just the absolute accuracy; it is how well the sensor can deal with large and fast-changing motions. In other words, the more money you spend, the worse weather you can survey in. Unfortunately, it is difficult to find information on how the sensor accuracy changes with rates of motion; most manufacturers give a single accuracy figure. I suspect that most sensors will give an accuracy that is better than the number they put on their specification sheets when there isn't much motion. We are planning comparative tests at some time in summer 2016.
There are quite a few motion sensors on the market to choose from. The "gold standard" is still probably the Applanix POS/MV series, at around €50k (Euros). Then there are various good units at around €20-30k that are OK for most marine situations. At around €15k, there is SBG-Systems Ekinox and SMC-S108. The SMC was one of the first sensors in that price bracket, and we have sold quite a few of them. You can see some motion artefact when the weather is bad, but you see little difference in the processed data product compared to using an expensive sensor. Ekinox is a more recent product, and we are very impressed with its features; for example, it is a true INS, giving corrected position as well as roll, pitch and heave, and the data is accurately time-stamped. We tend to recommend Ekinox for most users who can afford it, and I am interested to see how it's more accurate (and expensive) equivalent, Apogee, performs with Bathyswath.
Finally, there are a few interesting sensors appearing at around €5k, and this is the area that your question is addressing. These sensors tend to have a stated accuracy of 0.1 to 0.2 degrees. At 100m range, that means a depth error of 17-35cm; I suspect that most users could not tolerate that kind of error. But it is likely that the actual error when there is not much motion us much less than this, so the results could be acceptable. There are too many variables for me to make a firm recommendation for all cases, but I think that it is definitely worth trying. There are two sensors in this price range that I have seen recently that are worth looking at: SBG Systems Ellipse and Inertial Labs INS. Both of them include GPS as well as motion sensing, so are a good integrated option for small-boat survey work. The Ellipse has a good range of options of different data qualities, including dual-antenna GPS for heading. The Inertial Labs unit relies on a gyro-stabilised magnetic sensor for heading, which is probably OK for most applications, but not as accurate as the dual-antenna GPS. We have added the Inertial Labs data format to the Bathyswath software, and borrowed a unit to test it, but we haven't had the time to test it on the water yet. The Bathyswath software works with all the SBG sensor range.
Some of these sensors offer the option of post-processing the data to improve the motion accuracy; Inertial Labs suggest that 0.03 degrees can be achieved this way. I haven't tried this yet, but it sounds very interesting.

Add new comment

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.