Greetings all! I've got a project in the early planning stages, and I'm looking into some navigation options. I plan on competing in the Trinity College Fire-Fighting Home Robot Contest, and I want to use radar instead of an optical vision system.
I've found some tantalizing hints via Google about a radar chip that's been developed by CalTech, but I can't seem to find any info about it being commercially available yet.
Anyone have any leads?

 

I think any radar solution is going to be very difficult and expensive.

You can use radar techniques with ultrasound sensors which would be much cheaper and easier to get working. Since you will be working in a 2D plane you don't really need to get distances in the up/down directions. This will let you keep your sensors and drive electronics to a minimum. You can build a linear phased array of say 10 ultrasound transducers. By varying the phase of the drive signal you can simulate the effect of a lens allowing you to sweep the ultrasound beam back and forth with no moving parts. An FPGA or CPLD could generate all the phasing for your drive signals with a connected microcontroller computing the nessesary phase shifts.

 

The difficult part I'd agree with, that's why I'm starting research well in advance. But I believe the price of these chips was supposed to be quite reasonable.
Your phased ultrasound array is very interesting, I'll look into possibly implementing it. Thanks!

 

Reasonable cost really depends on who you talk to; if you're the military $10k each would probably be considered quite reasonable . For ICs low volume = high cost and radar is definately a low volume app.