Set membership approach to the propagation of uncertain geometric information (original) (raw)

1991, Proceedings. 1991 IEEE International Conference on Robotics and Automation

The fusion of geometric information is of great significance in multisensorial systems, mainly in robotics applications, where multiple sensors or mobile sensor systems that change their perspective of the environment capture sparse, and sometimes partial, geometric data. These data contain some level of uncertainty and, in general, some level of redundancy. Probabilistic approaches have been used to solve the problem of fusing this information to obtain the best estimate of a given set of parameters describing a collection of geometric features and its final associated uncertainty. Nevertheless, a probabilistic description of errors is not always available and only a bound on them is known. The setmembership approach postulates that a measurement only allows us to establish an uncertainty region in the space of parameters dcscribing a geometric feature. This approach avoids the general assumptions of unbiased and independent measurements taken by the probabilistic approaches.