The widespread application of Autonomous Mobile Robots (AMRs) in dynamic indoor environments has exposed the shortcomings of traditional navigation systems. For example, they cannot solve problems, like moving obstacles, layout changes, and multi-robot collaboration. The development of artificial intelligence and sensor technologies speeds up the changes in AMR navigation systems. This paper contains the development of AMR navigation systems. It analyses the technical challenges brought by dynamic indoor environments, including problems such as interference from mobile obstacles, bottlenecks in location and map updates, and efficiency conflicts in multi-robot collaboration. Additionally, it explores the key technologies used to solve these challenges, including dynamic object detection and removal in dynamic SLAM, multi-sensor fusion, semantic enhancement of dynamic map construction, and distributed task allocation based on a market auction mechanism, conflict search, and spatiotemporal path coordination in multi-robot collaborative navigation. According to these studies, this paper guides the development of more robust and adaptive AMR navigation frameworks to meet the application needs of modern indoor robots.
Shun Jiang (Fri,) studied this question.