<![CDATA[An omnidirectional robot moving in a crowd of robot soccer players requires autonomous navigation to navigate moving obstacles, generate safe, smooth, and adaptive trajectories when global information is unavailable. Global navigation and local control must be integrated so that spatial memory can balance speed, safety, and smoothness. This study proposes the implementation of Potential Field Obstacle Avoidance (PFOA) with a grid certainty map and a minimum histogram to address these challenges. This idea is based on the repulsive and attractive forces that shape the decision of the direction of motion in PFOA, which need to be accompanied by the confidence of the empty space map in the distribution of obstacle histograms that direct the robot to the right direction. Based on experiments that have been conducted, this approach has proven to perform better than the hybrid A* and bug methods. Our proposed algorithm is able to make the robot's travel time to the target consistently 13-21 seconds, with a smooth motion trajectory due to sharp maneuvers in the most difficult areas that are minimal, and the best safety confidence level value based on the certainty value between the two comparison methods. So it can be concluded that the strengthened PFOA is able to adapt to local dynamics and is superior in planning global robot trajectory maps that contain obstacles.]]>
Copyrights © 2026
