The Worldwide Journal of Robotics Analysis, Forward of Print.
This paper unifies the idea of consistent-set maximization for sturdy outlier detection in a simultaneous localization and mapping framework. We first describe the notion of pairwise consistency earlier than discussing how a consistency graph might be fashioned by evaluating pairs of measurements for consistency. Discovering the most important set of constant measurements is reworked into an occasion of the utmost clique downside and might be solved comparatively rapidly utilizing present maximum-clique solvers. We then generalize our algorithm to test consistency on a group-k foundation by utilizing a generalized notion of consistency and utilizing generalized graphs. We additionally current modified most clique algorithms that perform over generalized graphs to seek out the set of measurements that’s internally group-k constant. We handle the exponential nature of group-k consistency and current strategies that may considerably lower the variety of crucial checks carried out when evaluating consistency. We prolong our prior work to carry out knowledge affiliation, and to multi-agent techniques in each simulation and {hardware}, and supply a comparability with different state-of-the-art strategies.
发布者：Brendon Forsgren，转转请注明出处：https://robotalks.cn/group-k-consistent-measurement-set-maximization-via-maximum-clique-over-k-uniform-hypergraphs-for-robust-multi-robot-map-merging/
