Powered by EO: High Performance Sensing by Ocular Robotics

So, what we have got is a pair of ocular robotics systems. Right down here is called a camera system so it is building up a panorama using high pointing accuracy from whatever I give this, or you can just grab the images, drop them straight into the matrix. We don't need to do any image matching or anything like that. At the top here, we've got a 3D wide-eye system so that is producing the point cloud you see up in this section here, so that is doing a full scan of the whole area on frame. Down here you can see panorama, images coming up, depositing up to build ourselves a full shot and this lovely shot of our camera man right in the middle. Once we have got these two pieces of data we can actually combine them together, so we have very good information about the position where the images were taken, and very good information about the laser information. That gives us the ability to combine both of them and produce an RGB 3D point cloud. So that enables us to have both structure and color and texture information about specs. So, you can see here is the marching stand behind us colored in in red, after that we have got some more Northrop Grumman stuff so that is looking over in that direction. But you can see this is not just 3D data, you actually have got the color, and you've got the texture of the carpet, the texture of the floor in there. So that is the first demo we are running for you today. The second demo we are running for you is a basic security stall application just using the camera system. So, on this demo we are revisiting six target locations, two of them static. So, you can see we are actually revisiting this location to fix the level accuracy. You can actually see that image does not move at all. Simultaneously, updating static locations, we can also move our field of view. So, we can scan up and down, get multiple targets, scan across the sign up here, or even just look at something like a sign up here with varying camera exposure time as we expose the same location. So, we are effectively getting high dynamic range imagery. The Edmund Optics' material in this system: we have got five custom ground glass lenses installed to relay the optical information from the head down to the camera existing in the base. We also have Edmund's mirrors in both camera and the laser systems, and we used Edmund's off the shelf components to get to our prototyping stage, so we can produce a proof of concept of our camera system with adequate optical performance, but then for our full-on retail systems we are using all custom ground, custom design lenses.