Laser Triangulation for Navigation
I'm finally starting on this long-term project to determine robot position using an on-board scanning laser and a few fixed reflectors around the room. I'm curious if anyone else has implemented this successfully. I've done some preliminary experiments with the Hamamatsu P6986 sensor Philo used at http://www.philohome.com/sensors/lasersensor.htm Sensor works as advertised, range is ~6-7 feet even without a lens.
Naturally, the sensor is backordered at the only source, must find out if Junun will restock: http://www.junun.org/MarkIII/Info.jsp?item=79
Richard Vannoy 2001 interesting geometry document http://www.seattlerobotics.org/encoder/200109/lasernav.htm
Naturally, the sensor is backordered at the only source, must find out if Junun will restock: http://www.junun.org/MarkIII/Info.jsp?item=79
Richard Vannoy 2001 interesting geometry document http://www.seattlerobotics.org/encoder/200109/lasernav.htm
Comments
I've been meaning to try a laser pointer with my TSL1401 to see if it could pick up reflectors. The line scan camera is more general sensor and picking the right threshold should allow it to differentiate between ambient and laser light. It has the advantage of excellent tech support in this forum as well.
The TSL1401 doesn't take a complete image. It scans a single line of 128 pixels gray scale pixels. At first glance that doesn't sound that useful, but it's more useful than you would think as Robots mostly live in a 2D world. So you can tell relative brightness across the field of view and locate a bright light like a beacon or a candle flame. Phil Pi is the designer and has said that if you replace the lens with a vertical slit it becomes even more useful for this sort of thing.
Because it is a single black and white scan line the frame rate is good because you're not moving that much data around. Image processing of a single scan line is much easier to do as well.
I have also generated full color images by using colored LED's and stepper motors to move the target. The only thing I don't like about the TSL is the connectors... either a male plug should be provided or a reference to the cable. At the time I couldn't find one and ended up hand soldering my connections in a really tiny space:)
(sorry for the extra !!!'s but there is a 10 character minimum).
I looked for a maximum in the difference data between on and off laser states... I didn't have any intelligence in my set up. To be fully functional, you would have to do some data analysis to see if the surround is compatible with your expectations;.
Rich
http://www.junun.org/MarkIII/Info.jsp?item=79
If you like PhiPi's paper on BoeBot encoders and navigation, you'll love this! http://www.parallax.com/Portals/0/Downloads/docs/prod/datast/ApplyEncoder.pdf
If you are looking for Phil's paper on the BoeBot encoders, it's now located at:
http://www.parallax.com/downloads/applying-boe-bot-digital-encoder-kit