Labels
Posted by

Color Sensor Comparision

Philo has just put up a really well-done comparison of two of the color sensor that are currently available for the NXT. The first is the official one from LEGO, included in the V2 NXT kit, while the second is the HiTechnic V2 color sensor which can be purchased separately. I'm not going to bother summarizing Philo's excellent presentation too much, because reading it is an education in itself - not just on how the sensors "perform", but on how to compare sensors properly: developing tests, and understanding the results. Please read his page for the whole scoop. A dirt-simple summary is both have strong points, with the HiTechnic one being amazing at detecting multiple colors and handling challenging ambient light conditions, and the LEGO one (by virtue of using the rapidly-reading AtoD converter on-board the NXT) being much faster (probably important for line-following, for instance).

Thank you Philo for doing comparisons like this - and please read his summary, it really is excellent.
Labels:

Comments

NickNackGus said…
If anyone can contact Philo, could you ask him to add some graphs? It would help make it easier to decide which sensor to use under which circumstances. That and it would be interesting to look at any unexpected patterns.
Nov 16, 2009, 12:13:00 AM
Brian Davis said…
What type of graph would be helpful? For color sensitivity, it's essentially either "it does or it doesn't", while for speed it's just a time. In this case, I think the results are very well represented by a few simple numbers and a table to compare pluses and minuses (which he provided rather cleanly... much better than I was thinking of, actually). Additionally, if you really want to dig into patterns, he has his data linked (handy, that!)
Nov 16, 2009, 12:28:00 AM
Aaron said…
I want the graphs too for visuals. Another reason for the Lego Color Sensor is, you can make it flash a bunch of different colored lights! It can give people seizures! :D
Nov 16, 2009, 1:28:00 PM
NickNackGus said…
He did proviide all of his data, but it is spread out across several webpages, and I don't know how to change the formatting to make it into a spreadsheet.

Some things that could be graphed:
//Time required per sensor (bar graph)
//Max readable distance per color and sensor (surface; color is one axis, sensor the second, and distance the third)
//same as last but with different angle.
//Max readable distance per color and lighting condition (surface; color is one axis, lighting the second, and distance the third)
Nov 16, 2009, 10:04:00 PM
Unknown said…
Anyone know if the these sensors http://www.youtube.com/watch?v=xahQ6BMvri4 can be used in any of these projects?
Mar 5, 2010, 4:15:00 AM
Post a Comment

Popular Posts