作者: Sebastian Thrun
DOI: 10.1177/02783640122067435
关键词: Visual appearance 、 Randomized algorithm 、 Robotics 、 Robot 、 Mobile robot 、 Computer vision 、 Monte Carlo method 、 Computer science 、 Odometry 、 Artificial intelligence 、 Probabilistic logic
摘要: An efficient probabilistic algorithm for the concurrent mapping and localization problem that arises in mobile robotics is presented. The addresses which a team of robots builds map on-line while simultaneously accommodating errors robots’ odometry. At core technique combines fast maximum likelihood growing with Monte Carlo localizer uses particle representations. combination both yields an can cope large odometric typically found when environments cycles. be implemented distributed manner on multiple robot platforms, enabling to cooperatively generate single their environment. Finally, extension described acquiring three-dimensional maps, capture structure visual appearance indoor three dimensions.