The Mobile Robot Laboratory is engaged in long-term basic research in perception, control and planning for robots that navigate through complex indoor and outdoor spaces. Since 1981, building on work conducted at Stanford University since 1973, the lab has built three dissimilar mobile robots (Pluto, Neptune and Uranus) and demonstrated new methods that allow them to map and navigate cluttered surroundings by stereo and monocular vision, by broad beam sonar, scanning laser rangefinder and other sensors.
The two decades of experience have given us the means and confidence to develop a mobile robot control program with enough spatial competence to reliably execute tasks like delivery and floor care in normal areas, with no special navigational aids and trained only by an initial run through. Our current efforts may result in a laboratory prototype of the first mobile robots with mass market potential.
Our approach accumulates sensed evidence about surrounding geometry in a spatial grid. We plan to integrate about ten thousand 3D location estimates from multi-baseline stereo vision several times per meter traveled. Preliminary, but very efficient, implementations of stereo and 3D grid algorithms suggest that 100 to 1000 MIPS of computer power will suffice to safely guide walking-speed robots through normal indoors. This amount of computing power will be inexpensive by the end of the decade, and we anticipate the widespread use of freely roaming robots shortly thereafter.