Water Linked DVL A50 Support

[EliotBR (Eliot)]

Hi @yuki,

@yuki

On the DVL, “Velocity valid” is valid, but Ardusub’s DVL driver shows “timeout restarting”.

In the air, “valid velocity” is invalid, but Ardusub’s DVL driver shows the status as “running”.

When submerged in water, “Velocity valid” is enabled, but the status becomes “timeout restaring”. (If only the DVL is submerged, “Velocity valid” is enabled and the status remains “running”.)

Interesting. I’m not sure whether that’s an issue with the driver (which seems unlikely to have not been picked up before now), or just some unfortunate timing when you happened to be checking each component.

A couple of questions:

1. Was this on BlueOS or Companion, and which version?
2. Is this behaviour repeatable (e.g. every time you remove it from the water the velocity goes invalid and our driver changes status from “timeout, restarting” to “running”)?

@yuki

Does this mean that “Velocity valid” is invalid in failure cases?

My description was for cases that may cause any DVL to capture sufficiently poor data that it can’t accurately determine the velocity. Whether it detects those conditions correctly depends on the particular sensor implementation (both hardware and firmware) - i.e. an ideal sensor would always correctly detect and report when its readings are and are not valid, but some sensors may fail to detect certain invalid situations (e.g. if the data seems valid but actually isn’t, which could occur naturally or by some intentional means to trick the sensor).

@yuki

1. interference between the Ping Sonar 360 and DVL sound waves

The Ping360 uses a 750kHz transducer, whereas the DVL-A50 uses 1MHz. It’s unlikely (but not impossible) for them to have some interference, depending on the receiving hardware and logic on each. You could quite easily test if the Ping360 is causing issues by turning it off / not connecting to it and seeing if that makes the DVL start working properly.

@yuki

2. a mismatch between the IMU in the Ping Sonar 360 and the IMU in the DVL or ROV.

The Ping360 has no IMU, and does not communicate position or orientation, so that’s not possible. It is possible for there to be a mismatch between the DVL and ROV IMUs and/or between their predicted positions and orientations (which is discussed a bit here), but that would exhibit as an issue of unreliable position estimates, not something reported directly by the DVL itself or by our driver for it.

Note that our DVL driver does not currently report orientation at all.