부산대학교 도서관

The application of autonomous robots has been increasing in agriculture sector to substitute human labour and to improve the production yields. A self-sufficient robot is intended to accomplish specific jobs in different locations of the working field area, thereby an economical and effective navigation system for differential wheeled mobile robots is a paramount importance. In this paper, an autonomous navigation system of an agricultural mobile robot is proposed using pure pursuit algorithm (PPA) which is also assisted by vector field histogram (VFH). PPA autonomously guides towards waypoints, whereas the VFH algorithm helps the vehicle steer away for obstacles. The 2-dimensional light detection and ranging (LiDAR) sensors are used to monitor through the VFH algorithm. Minimum number of waypoints are set in PPA for convenience on map setup. Several indicators such as distance covered by robot, number of iterations required for completion of travel, etc., are investigated with the variable settings in PPA algorithm. Result analysis shows that mobile robot can travel at speed range of 2.5-25 km/hr with obstacle evasion which is adequate for agricultural mobile robots.