抄録
This paper presents the result of preliminary study toward applying Simultaneous Localization and Mapping (SLAM) to the hydraulic-driven 6-legged robot, COMET-IV. However the SLAM concept to be applied is a different approach compared to the method applied by other researchers. We are proposing a method that is utilizing the good points of Occupancy Grid Mapping method (OGM) and Extended Kalman Filter (EKF) method, which could optimize data processing time and produce precise localization of the robot in outdoor environment. The OGM is used to generate a shortest and safe travel path from starting point to the targeted goal, at the beginning of the journey and the EKF is used to simultaneously updating the robot's location while moving toward the targeted goal. All the data used to determine the coordinates of the robot and the obstacles are based on 3D points cloud data produced by rotating laser range finder (LRF) together with gyro data and the robot's center of body (COB) data. The experiment shows promising result.
