[Dprglist] Assisting PID odometry with an IMU

[Murray Altheim]

On 24/01/21 12:32 pm, markus wrote:
> Hi Murray,
> 
> I'm not sure about your architecture, so this might be redundant. Many
> years ago we had a similar discussion here and as a result I
> implemented 3 PIDs for my differential steered robot.

Hi Markus,

I think Doug Paradis or Jian somebody mentioned this awhile back on one
of the videos, basically that you can use a PID controller as controlling
the differential for the steering (or as David A. would say, control of
theta). It sounds like a great idea if one is controlling a robot with
velocity+theta rather than left_power+right_power, which is my aim
(eventually). So this fits in with my architecture just fine.

> One PID (A) does the steering and determines the set points for each
> motor. And those are used as the input values for two PIDs (L and R),
> one for each motor. L and R use the same configuration and ensure that
> each motor actually does what is expected of it, without impacting the
> other motor. Note that L and R should run faster than A. I have A
> running at 100Hz and L/R running at 1kHz - I chose 10x more or less
> arbitrary.
Thanks for adding in the notion of running the A PID slower than the
L/R PIDs -- that sounds like good advice. I note that you're running
your PID loops (A, L/R) a *lot* faster than me, as I've been doing
everything at 20Hz. Hmm.

Cheers,

Murray

...........................................................................
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