Military Robots Get Real Depth Perception on the Way to Autonomy

Most of us understand how conventional 2D video camera sensors work, and we know how lasers can be used to determine distances to a target. What if we could combine the two technologies? If a laser pulse could be split optically into a grid, say 128 by 128, and used to scan a target so that the delay times of the reflected pulse 'pixels' tell us the distance to the object, then we might have an extremely useful 3D scanner. And, if we could repeat the process fast enough, say 30 fps, then the resulting scans could be used as part of the sensor and control system for next generation autonomous robots.

According to a recent report in Design News, that's the approach being explored by iRobot Corporation to design their next generation military robots. They're attempting to give the PackBots robots improved real-time depth perception using 3-D Flash Ladar sensor systems developed by Advanced Scientific Concepts, Inc

Although the sensor's basic principles are fairly straightforward, actual implementation can become rather complex and potentially expensive. Don't expect to see it rolled out at the consumer level with Roomba like products in the near future. However, for dangerous or demanding applications, like assisting or replacing military personnel in the field, it has tremendous potential.


Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>