Java’s sleep resolution goes to 1 millisecond - there are nano-sec - exotic calls, but they have iffy implementations (ie don’t work)
Also, the sleep means the scheduler will put it to sleep for that long and at the end of that period, mark the thread ready, but that does not mean it will run… And that is why real-time operating systems are around… go figure…
Wonder how much problems rounding to the nearest millisecond will cause? Well, I guess I know where to start… 300 baud… brings me back to the days of handsets
Have you tested the wiidar outdoors? I am thinking of entering the http://www.rie2011.org/robotour/ competition and I am researching sensors, mostly for obstacle detection.
Didn’t get the opportunity, before I fried the wii’s regulator with too much power. It was pretty robust indoors. For a short range outdoor system it would probably work. You could add more filtering for 650 nm if your going to use a cheap key chain laser. You could also use a polarized filter for the laser, I thought this could add quality filtering - since all the light from any laser is polarized in the same way, just have to make sure the laser and filter are oriented in the same way. Higher wattage IR Laser would probably extend the range considerably, but the added safety concerns might not be worth it.
I was meaning to get another one, and make an incredibly easy to build and powerful mapping robot. It would basically be a wiimote with a motor controller. You could drop it on the floor and it would map your whole house.
I had a computer talking to it using wireless calls to the LEDs as a form of 3 bit parallel port. The idea being the computer would map everything and tell the little tiny bot where to go. I think it has potential, but have been too busy with other stuff to get back to the project. I fried it when I did not have 3.3V motor controller and overpowered it with 5V.
If you do get a wii controller, you might be interested in MRL’s WiiDar service to connect to and visualize your Wii data.
