General Competition (April 1992)

Autonomous Navigation in Quasi-Structured Environments

Sponsor: Pyxis Corporation (formerly HelpMate Robotics)

• Project Performance Period: 6/15/1992 - 3/31/1994
• Total project (est.): $1,136,235.00
• Requested ATP funds: $698,596.00

Material handling is a major cost in manufacturing, service, and distribution industries. However, automatic guided vehicles (AGVs) -- self-navigating robots -- have found only limited use in industry and commerce despite myriad potential applications for parts delivery, building maintenance, security, and similar simple tasks. A major problem is that (with the exception of multi-million dollar space probes and experimental military devices) current AGVs require fixed navigational structures -- a wire guide buried in a floor, for example -- and cannot navigate and avoid obstacles reliably in the largely unstructured environments typical of large factories, warehouses, distribution centers, and other places which otherwise would be logical customers for an AGV. TRC proposes to develop a combination of technologies that will make possible affordable and reliable automated navigation in such quasi-structured environments. Planned elements for this advanced AGV system include a low-cost Light-Direction-And-Range (LIDAR) scanner (today's military and space LIDAR cost upwards of $50,000), navigation based on observing features on the ceiling, and improved algorithms based on artificial intelligence concepts for integrating multiple sensor inputs and navigating obstacles.

For project information:

Joseph F. Engelberger, (203) 798-8988

ATP Project Manager

Jack Boudreaux, (301) 975-3560
jack.boudreaux@nist.gov