Safe mobile robot navigation in human-centered environments using a heat map-based path planner

摘要

Safe robot navigation in human-centered environments is important to avoid collisions. A major limitation of the traditional path planning algorithms is that the global path is planned only with the knowledge of static obstacles in the map. This paper presents a novel 'HMRP (heat map-based robot path planner)' which uses fixed external cameras to generate a heat map of different passages based on congestion, so that robots can generate congestion-free paths at the global planning stage itself. The congestion values are maintained in a database and the paths are classified into hot and cold regions. Robot navigation is affected by the direction of movement of people. Hence, in this work, the HMRP-based planner also considers the direction of movement of people in passages which improves robot navigation. The proposed HMRP is compared with traditional path planning algorithms in real environment. Results show that the proposed HMRP algorithm generates congestion-free paths for safe robot navigation.