作者: Randall C. Smith , Peter Cheeseman
DOI: 10.1177/027836498600500404
关键词:
摘要: This paper describes a general method for estimating the nominal relationship and expected error (covariance) between coordinate frames representing relative locations of ob jects. The may be known only indirectly through series spatial relationships, each with its associated error, arising from diverse causes, including positioning errors, measurement or tolerances in part dimensions. estimation can used to answer such questions as whether camera attached robot is likely have particular reference object field view. calculated estimates agree well those an independent Monte Carlo simulation. makes it possible decide advance uncertain accu rately enough some task and, if not, how much improvement locational knowledge proposed sensor will provide. presented generalized six degrees offreedom provides practical means esti mating relations...