Subject:
|
Re: Balancing robots
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 28 Jan 2004 05:55:58 GMT
|
Original-From:
|
Gordon Elliott <gelliott@csisc.%ihatespam%cc>
|
Viewed:
|
1158 times
|
| |
|
|
Has this link been given:
http://robotics.ee.uwa.edu.au/theses/2003-Balance-Ooi.pdf
(PDF file)
Rich Chi Ooi's "Final Year Thesis" in mechanical engineering (U of Western
Australia) on a balancing robot. Mentions Steve Hassenplug's Legway.
I'm convinced that Peter is on a useful track (in that wheel rotational
speed information can successfully be gathered from back EMF). However this
will not give the robot's tilt, which must additionally be known. So this
can only be part of the necessary balancing information and the tilt
information problem must also be solved before knowing how useful the back
EMF as whell velocity will be.
PS, I've got some more ideas on circuits to derive the back EMF which I will
explore when I get more time later in the week, if anyone is still
interested and has not solved all the problems already. Peter I'll reply on
some other topics when I get a little more time, as I am very interested in
this problem (from theoretical standpoint).
The web sites in my previous e-mail show why the special turn-off cycle idea
is a good one, in that the "off" case voltage cannot be measured during the
regular function of the PWM because the current typically never drops to
zero during a cycle (except for crossing zero in a triangle wave if
sufficiently near zero average current). One of the web sites shows the
protection diodes, and even when turning off the transistors (the "on or
open-circuit" control version) the diodes conduct until the inductive
current reaches zero. That means that the circuit is not really "open"
during the supposedly "open" PWM cycle. That is why one has to do the extra
settling delay, even with a configuration that opens the main transistors --
or use a different method.
I had never heard that idea was in use, I just realized that had to be done
after coming to recognize how the PWM suppy/driver acts. But I am also
thinking that dynamic measurement with an op amp circuit during normal PWM
function will not be very difficult. One could either use the CPU to do the
subtraction and just measure motor current, or do a fancy circuit that
actually subtracts the drive "locked rotor" current equivalent, to find the
equivalent "back EMF" current. There is no problem with the PWM, just
average over the cycles with a capacitive circuit, as long as one can pick
the current information out.
-- Gordon
|
|
Message is in Reply To:
 | | Re: Balancing robots
|
| Gordon (...) Good descriptions but a little basic for what we're discussing here. PWM can be used to control balancing in three ways. In the two parts of the cycle, the transistors are: - on or open-circuit - on-forwards or on-backwards - (...) (21 years ago, 27-Jan-04, to lugnet.robotics)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
