Distributed pose graph optimization method and device based on multi-level graph segmentation

• Main Authors: LEE JON-HO, GUO GUANGHUI, HONG YIGUANG, YI PENG, LEI JINLONG
• Format: Patent
• Language: Chinese; English
• Published: 15.09.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention provides a distributed pose graph optimization method and device based on multi-level graph segmentation, and the method and device are used for the communication among a plurality of robots, and the method comprises the steps: carrying out the multi-level graph segmentation operation of an initial pose graph, and constructing a distributed optimization problem; updating the selected variable block by adopting a Riemann gradient coordinate descent algorithm until an optimal solution meeting conditions is searched; and projecting the optimal solution to an attitude graph optimization feasible solution by using singular value decomposition. According to the distributed pose graph optimization method based on multi-level graph segmentation, a balance optimization sub-problem which is easier to process is constructed, the communication cost in a multi-robot network is reduced, and the overall optimization performance of a multi-robot system is improved. 本发明提供一种基于多级图分割的分布式位姿图优化方法及装置，用于多个机器人之间的通信，所述方法