Answers are grounded in this paper and cite their sources.
A new laser vision system is proposed for the navigation of mobile robots in pipe environments. This paper mainly focuses on forward recognition of individual T-junctions and elbow pipes to prevent the robots from becoming stuck in pipes. We propose to mount weaving 2-D laser scanners on robots to obtain selective 3-D front-pipe maps. Recognition of T-junctions requires 3-D maps for pipe wall on wheel sides, whereas recognition of elbow pipes requires full-forward 3-D maps. From these selective forward 3-D maps, algorithms for recognition of T-junctions and elbow pipes were developed, utilizing the mathematical approach to describe the pipeline structures. The practicality of the proposed algorithms was validated for several pipe environments, including T-junction, single elbow pipe, and the latter pipe of double elbow pipes.
No takes yet. Share an insight, caveat, or question.
Kim et al. (2018) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: