Subject:
|
Re: sonar for obstacle avoidance
|
Newsgroups:
|
lugnet.robotics.handyboard
|
Date:
|
Wed, 1 May 1996 08:18:49 GMT
|
Original-From:
|
John Wasinger <jwasinge@ichin%stopspammers%.Mines.EDU>
|
Viewed:
|
1510 times
|
| |
|
|
You mean besides making it 'crawl'? In my robotics class were were
programming autominus vehicle equipped with a servo mounted sonar on the
front. The basic process is this:
1) scan
2) analyze data... (aka look for wall shaped thing)
3) adjust coarse by "repelling away from the wall"
4) move
We had the robot follow a path defined by a flowing vector field. When
it detected an obstacle it applied a repelling flow force emanating from
the region of the object. We used a exponential repulsion field based on
the distance from the obstacle to the robot so as to weaken its repulsion
if the robot was far enough away. We added and subtracted these repulsion
fields from our original path field so as to keep the robot on coarse.
It's been a year since I took that class though... so some of my
terminology is a bit rusty.
From what I recall we never stopped the motion of the robot. We Just
slowed it down to account for the response time of the sensor sweep
analysis.
I'll go through my old class notes and give you a more detailed
response later if you want.
Hope I helped...
John Wasinger
EE CS major at Colorado School of Mines
I need to make my sig file more flashy...
|
|
Message is in Reply To:
 | | sonar for obstacle avoidance
|
| Ok, it seems that everyone thinks the Polaroid sonar module is pretty nifty. However, how is it for _obstacle avoidance_? My new robot has a dire need for a way to keep from crashing into things all the time, and sonar would seem to be a nice (...) (28 years ago, 1-May-96, to lugnet.robotics.handyboard)
|
2 Messages in This Thread:
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
