抄録
In this research, we propose the system that moving objects which are in a robot's circumference using a Laser Range Finder (LRF) is extracted, and those numbers and position coordinates are estimated. The data of the object of the circumference obtained from LRF is expressed as existence probability on a grid map for the ease of carrying out of noise reduction or processing, and moving objects are extracted out of surrounding objects by performing difference to it. Moreover, the number and its position coordinates of the moving object which performs clustering by the MeanShift Method and exists in the circumference from the Occupancy Grid Map after performing difference in order to estimate the number of a moving object are presumed. We demonstrate the moving objects detection for the purpose of showing that the proposed system is more valid than using raw LRF data.
