[Dprglist] Optical flow sensors - down the rabbit hole

Murray Altheim murray18 at altheim.com
Fri Jul 3 04:48:04 PDT 2020

Hi Karim,

I'm impressed with your investigation and must say thank you for going to the trouble
of going down the rabbit hole with me, it's much appreciated. I can give a bit of
feedback to you regarding my communications with Pimoroni: my message regarding the
PAA5 has been forwarded to their engineering team, with "fingers crossed" from the
guy I'd been chatting with.

I'm wondering if you or me should inform the Pimoroni team about the ground plane issue
or do you think they'd figure it out on their own? And I'm wondering if we'd just be
confusing them to mention the PMT9101DM-T2QU? OTOH, their Breakout Garden series could
use a few new sensors, particularly those tailored to use specifically on robots. I
keep thinking of asking them to pop a Sharp IR sensor on one, as that'd be handy.

As for the suitability of the PAA5 for very fast robots, I can only say that this would
not be an issue for anything except quite speedy robots as you note. For mine the
detection speed would be fine: none of my robots could get to anywhere near 45 inches
per second unless it dropped off a cliff.

I'd also add that my current plan (which changes all the time) is to use such a sensor
on a Mecanum robot as a *supplement* to the encoders and IMU, not solely as a means of
odometry. We discussed the issue of using two of these at a distance around the robot's
central axis as a way to track rotation-in-place, but frankly that's a lot of extra
complexity and space usage for a small robot so I'm likely to stick with one sensor and
just note that if the program is rotating the robot in place I'd take that into account
in the odometry measurement. I.e., I'm thinking the real value is when the Mecanum
wheels are doing something relatively complex, uh, like not rotating in place. Hmm.

    [he says speaking as if he had any idea what he's talking about...]

But I guess the main point of this discussion is a confirmation that yes, we'd love to
have an alternative optical flow sensor board based on a more suitable sensor, rather
than a sensor meant for drones repurposed to <10cm range.

It's pretty late here so I'll have to read up on the sensors over the weekend. Thanks
for the research and references!



Murray Altheim <murray18 at altheim dot com>                       = =  ===
http://www.altheim.com/murray/                                     ===  ===
                                                                    = =  ===
     In the evening
     The rice leaves in the garden
     Rustle in the autumn wind
     That blows through my reed hut.
            -- Minamoto no Tsunenobu

More information about the DPRGlist mailing list