Subject:
|
Re: Multitasking
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Fri, 14 Jan 2000 19:06:19 GMT
|
Original-From:
|
David Leeper <david.leeper@destinysoftwareNOSPAM.com>
|
Viewed:
|
712 times
|
| |
|
|
Hi Alex,
I was thinking this over last night and was wondering about how sleeping a task
would affect an application that requires precise timing. In the bicycle
example, what would happen if the balancing task was asleep when the bicycle
began to tip?
I think multitasking is certainly useful in situations where the tasks are
disjoint, such as driving the robot and sending a message over the IR port. But
when the tasks need to work together closely, the sleeping issue worries me.
David Leeper (thinks this is an interesting topic)
alex wetmore wrote:
> From: "David Leeper" <david.leeper@destiny.com>
> > I removed the multitasking features from my cruise and avoid code last • night
> > and discovered the robot runs smoother, using less code, and less RCX
> > variables.
>
> Cruise and avoid is a pretty simple type of program, and multitasking really
> isn't necessary. I wrote a sublimation version of it mostly to play with
> the concepts of sublimation. Two nights ago I wrote the same program using
> LegOS and used a more traditional loop there. The two versions are:
>
> http://www.phred.org/~alex/lego/wander.nqc
> http://www.phred.org/~alex/lego/wander.c
>
> I think that the sublimation model becomes a lot more powerful as you have
> many inputs to monitor and build more complex robots. My cruise and avoid
> robots really only have one input to watch -- some sort of bump detection (a
> rotation sensor in the above examples).
>
> An example of a robot type where I think that sublimation would be useful is
> trying to build a balancing and steering vehicle such as a bicycle. Alex
> Nicolaou and I started talking about this in private email a few days ago.
> Here you might want a high priority task which makes the minor steering and
> weight-shift movements necessary to keep the bike balanced, while having a
> lower priority task which controls the same steering and weight-shift
> movements to allow the bike to turn. We'll see how this works out once I
> have time to build such a robot.
>
> Conclusion: multitasking isn't necessary or optimal for all tasks, but it
> can be useful in more complex tasks.
>
> alex
|
|
Message is in Reply To:
 | | Re: Multitasking
|
| From: "David Leeper" <david.leeper@destiny.com> (...) night (...) Cruise and avoid is a pretty simple type of program, and multitasking really isn't necessary. I wrote a sublimation version of it mostly to play with the concepts of sublimation. Two (...) (25 years ago, 13-Jan-00, to lugnet.robotics)
|
9 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
