I’ve been thinking about this for a while, and I am starting to pull it together now.
With the RPi and Coral I’ll have general processing with AI/ML acceleration. I’ve run the image recognition demonstrations, and it takes a program that runs in 450ms and makes it run in 4ms. The GPS-IMU provides GPS, and 10DOF position information (x, y, z, acceleration, magnetic heading, pressure altitude, and temperature). So, a super fast pocket sized computer brain that can sense it’s environment and send signals out to the world.
Time to dust off my robot law books. More to come. . .