Ping2 performance under turbulent environments

General Discussion
1

Hello, we are using the Ping2, as an obstacle detection sonar on a USV, and is currently mounted at 20 degrees to the horizontal. When performing different static and field tests, we noticed that when the sensor sees turbulence(for example, disturbance caused due to the USV hull, motors, etc), the confidence and distance outputs, are complete garbage. I was wondering if there is a way to fix this. We have looked at the other post regarding mounting the sensor at an angle and understand the limitations that come with mounting at an angle. I also understand, that turbulence causes issues with data, and are looking for suggestions to improve it.

The following are the results from our static and field tests, each image is a result of a scenario, as explained below

  1. Image 1: This was performed under a static bench(Water Tub(4’ x 4’)) and provides a baseline to what is expected under normal conditions.

  2. Image 2: In this static test case, a jet of water was initiated at one end of the tub. We already see a small deviation in data, still within an acceptable range

  3. Image 3: In this static test, a lot of turbulence was generated, away from the sensor. The data at this point is unusable

  4. This is a field test, where the USV was moving at its maximum speed. Again unusable data

Image 3
Image 31000×1000 59.6 KB

Image 2
Image 21000×1000 41.5 KB

Image 1
Image 11000×1000 30.1 KB

image 4
image 41000×1000 100 KB

2

Hi @dskumar -
Unfortunately, when any sonar encounters bubbles and large disturbances in the water, reflections can cause issues in determining the acoustic response. I would think the sensor should be downwards facing to monitor depth (helps avoid running aground) and minimize issues with low-confidence readings… For collision avoidance, a sensor intended for use in air like lidar or ultrasonic range sensors may make more sense?
If you’re set on using the Ping2, it could be mounted horizontally or at some forward-looking angle, and at a depth so that it doesn’t encounter surface affects? This obviously has range impacts.
I’d also expect the sonar to perform much better in large bodies of water, without flat, sound-reflective walls in close proximity. Tanks and pools have significant multipath reflection issues!