Currently, many of mobile robots are stand-alone. The robots acquire and utilize sensor data and map information by themselves. Sharing of sensor data and map information by networked multiple robots will be efficient for recognition and decision making in large scale environments. This paper introduces a map sharing system based on block chain for mobile robots. In the proposed system, multiple robots share Human Frequency Maps which include human existence frequency in each grid of the maps human. Especially, this paper focuses on updating Human Frequency Maps based by multiple mobile robots. We developed a simulator for map sharing and updating. As a result of simulation, map sharing and updating can be achieved, based on evaluating observation number and distance to each grid.
抄録全体を表示