Optimal sensor fusion for land vehicle navigation

Technical Report · Mon Oct 01 00:00:00 EDT 1990

DOI:https://doi.org/10.2172/6242453· OSTI ID:6242453

Morrow, J D

Position location is a fundamental requirement in autonomous mobile robots which record and subsequently follow x,y paths. The Dept. of Energy, Office of Safeguards and Security, Robotic Security Vehicle (RSV) program involves the development of an autonomous mobile robot for patrolling a structured exterior environment. A straight-forward method for autonomous path-following has been adopted and requires digitizing'' the desired road network by storing x,y coordinates every 2m along the roads. The position location system used to define the locations consists of a radio beacon system which triangulates position off two known transponders, and dead reckoning with compass and odometer. This paper addresses the problem of combining these two measurements to arrive at a best estimate of position. Two algorithms are proposed: the optimal'' algorithm treats the measurements as random variables and minimizes the estimate variance, while the average error'' algorithm considers the bias in dead reckoning and attempts to guarantee an average error. Data collected on the algorithms indicate that both work well in practice. 2 refs., 7 figs.

• Research Organization: Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
• Sponsoring Organization: DOE/NE
• DOE Contract Number: AC04-76DP00789
• OSTI ID: 6242453
• Report Number(s): SAND-89-0816; ON: DE91002586
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE
ROBOTS
NAVIGATION
ALGORITHMS
COORDINATES
DISPLACEMENT GAGES
ERRORS
MONITORS
SECURITY
MATHEMATICAL LOGIC
MEASURING INSTRUMENTS
420200* - Engineering- Facilities
Equipment
& Techniques
990200 - Mathematics & Computers