Just been watching a video on one of the other channels I subscribe to "Applied Science", and he was trying to make a non-reflective black surface. One thing the guy did say is that black velvet only has a reflection value of 0.4% which would make it ideal for calibrating the colour sensors and that white paper only reflects 90% of the incident light. Therefore, there could be room for adjusting the values during white calibration down to 230,230,230 to compensate.
Another way to make a near-black surface would be to hold a piece of cold metal above a candle flame to deposit carbon on to the metal surface.
Thanks to Bill's recent video I just ordered 5 of these from the other end of the silk road. ?
I've actually been thinking of using these on my robot as a way of verifying that it has made it to a particular location. I've been thinking of making round color patches that can be fasted to the floor in specific places. This way when the robot gets to that position it can verify it by the color that is underneath it.
I thought that I might also use small color patches on the walls as position identifiers too. In that case I'd need to have the color sensor be capable of moving from looking downward at the floor to also being able to look horizontally at the wall. So I might need to employ a servo to move the color sensor between these two position. Either that or just use two color sensors which would probably be both easier and cheaper than employing a servo to move a single sensor around.
In any case using color sensors as a means of detecting whether a robot is at a particular location is the basic idea.
white paper only reflects 90% of the incident light
I'm wondering why a mirror couldn't be used for the white values? A mirror would reflect back the exact same white light that is coming from the source LEDs. Seems like this is what should be desired for calibration. Then white paper may actually read a bit lower.
DroneBot Workshop Robotics Engineer
James
This way when the robot gets to that position it can verify it by the color that is underneath it.
That had crossed my mind too!
But I had only got as far as recognizing the floor colour of the room the robot was actually in. I have only ordered one sensor and it arrived 30 minutes ago, sourced locally due to impatience!
My single ATmega328P with bootloader also arrived 30 minutes ago, so I guess that is going to keep me busy for a few hours!
I have only ordered one sensor and it arrived 30 minutes ago, sourced locally due to impatience!
I solved the silk road instant gratification problem. I simply order something new every day.This way there's always something arriving pretty much daily from across the globe. And a lot of it is a very pleasant surprise because by the time things get here I usually forget what I had ordered. ?
Like yesterday I got a fairly large package in the mailbox. I couldn't imagine what it could be as most packets are pretty small. I opened it up and found it to be filled with 3 different lengths of DuPont wires. What a pleasant surprise. I had completely forgot that I had order those.
DroneBot Workshop Robotics Engineer
James
Just been watching a video on one of the other channels I subscribe to "Applied Science", and he was trying to make a non-reflective black surface.
So if the colour's: red, orange, yellow, green, blue, indigo and violet make up white light, will their mixed inverse colour's (or a mixture of colour's) cancel each other out to create a true black light? 😀
Good channel btw... seen a few video's in the past, always interesting stuff!
I guess that you will only get black by mixing the complementary colours! ? ?
But this more a question for James Clerk Maxwell as he discovered the RGB mix!