I think the way to look at it is that this approach presents a whole new way to use LIDAR. That "puck" has a range of 100m, weighs a pound and is designed to spin through 360 degrees. The LIDAR-on-a-chip shown here has a range of 2m (potential to 10), weighs effectively nothing and has a FOV of 50 degrees. It also needs a laser and some collectors so it's not a complete system but obviously yeah, it's hella cheaper. My guesstimate is that for navigation you're still gonna want Mr. Spinny Turret or an iteration thereof. But this application of LIDAR would actually allow you to put a couple on your waldoes for manufacturing manipulation and stuff.
I didn't catch that on my read. I'd imagine that's pretty limiting then. Even if you coated your car in these and had über-machine-vision reconstruct the 3D landscape around the car, you'd still not be anywhere close to catching objects at the edge of your stopping distance. That can barely work for Google's tiny car let alone a vehicle going 60 mph on the highway.The LIDAR-on-a-chip shown here has a range of 2m (potential to 10)
Was it you that linked to the waveguide approach to LIDAR some time in the past six months or so? The advantage of doing it that way is it's cheap as fuck and your slew rate is in the MHz range. The disadvantage of doing it that way is you're limited to super low power lasers which means you're going to be limited in range by pure physics. Thing of it is, though, it opens LIDAR to new applications. Who knew 30 years ago that it'd be nigh onto impossible to buy a car without ultrasonic proximity sensors in the bumper nowadays? Being able to get high-resolution 3d topology on anything from 2m away with a $30-$40 (retail) component starts to make shit really fucking cool. I'm no booster of 3D printing but theoretically you could wave your sonic screwdriver over a key, upload it to your glowforge over bluetooth and dupe that fucker out of sheet aluminum with a button press. Or wave it over your mom's broken vase chips, flip them over, do it again, and have your 3d printer derive the piece that you can't find. There are some cool applications for cheap-ass spacial imaging.
Yep 1, 2. I guess there's multiple ways to engineer it? "Lasers bounce off stuff" is near my limit on my knowledge on these things.Was it you that linked to the waveguide approach to LIDAR some time in the past six months or so? The advantage of doing it that way is it's cheap as fuck and your slew rate is in the MHz range. The disadvantage of doing it that way is you're limited to super low power lasers which means you're going to be limited in range by pure physics.