False alerts RLT515 low speed ride

It happens not depending on location. Even on "open area".

I updated firmware to newest version, but still the same issue. It's quite strange, because higher speeds, no issue and radar detect cars properly.

Any idea?

Hmm yeah sounds like something is not working right then, as it definitely shouldn't be triggering randomly when there are no objects there. Might be best to contact Garmin Support in case it's defective.

It's already over 2 years old. Anyhow I'll ask Support. Thank you for your advice. 

"This is to be expected because the signal path between the radar and the object is actually shortening (at least briefly) as you pass it, due to the geometry"

Maybe it's too long since I did my physics degree but I don't think this can ever be the case? The closest approach will surely be when you're at 90degrees to the object, with it directly to your side, and from then on its distance will continually increase? If you're going in a straight line.

Hi, we don't know the sensor's detection processing algorithms, so my comment is what I infer from multiple repeatable examples of false positives.

Relative motion definitely seems to be one, such as when the bike is turning (or levelling out at the bottom of the steep descent) and the radar "beam" is being swept across objects. In one case I'm sweeping through a turn in front of a funky angular-styled building. It appears that the geometry is such that the path between the radar and the building is shrinking as I pass it. In another case I'm turning at a specific intersection and some object there (not sure what exactly) sets it off as the beam sweeps across it.

The other cases seem to involve compound surfaces. The signal could be bouncing between 2 or more objects before returning to the radar (so a complex return path rather than simple out-and-back). Or just simply giving multiple returns from different surfaces. Some examples:

- Passing a corrugated metal wall with a round lamp post next to it. Only the post seems to set it off, but I speculate it's the combination of the post and wall that does it (since neither isolated posts nor walls without posts seem to trigger it). I've tried to think thru the dynamic signal path as I pass it and the headache makes me drop it

- Entering a bridge underpass or a tunnel. The entrance sets it off, and I think the false-positive clears before I exit the tunnel (if it's long enough). There are repeating surface shapes on the walls/ceiling. If the radar detects the first one, the second one could give a similar return as you move forward, etc. So I speculate it creates a strong enough target at a certain distance (like it's "following" you) for the algorithm to say hey there's an object there (at least briefly). Whether briefly sensing a closure rate is necessary or not, I don't know.

- Passing a certain rural mailbox. I assume it marks the entrance to a secret govt lab lol

In any case, the how/why doesn't particular matter. Suffice it to say false positives can occur, likely in repeatable situations. (I've yet to see any reports of false negatives, which is good!). However if false positives are occurring frequently where there are no objects and can't be repeated, which might be the OP's case, then perhaps something is wrong with the unit.

Thinking about the physics is certainly interesting tho. I mean it's pretty amazing how such a small unit can work so well with all the clutter out there. Cheers!