Obstacle detection for intelligent robots based on the fusion of 2D lidar and depth camera
Bailin Fan, Hang Zhao, Lingbei Meng
Abstract
To address the limitations of traditional obstacle detection methods that rely on single sensors and cannot accurately detect and locate obstacles in complex environments, this paper proposes an obstacle detection method based on the fusion of 2D lidar and depth camera. The proposed method converts the data from the two sensors into lidar data in the same coordinate system for clustering analysis and obstacle identification. It uses Kalman filtering to estimate and predict the target state, significantly improving the range and accuracy of obstacle detection and providing more reliable obstacle information for intelligent robots. Experimental results show that the proposed method outperforms other commonly used methods in actual indoor scenes, demonstrating that the fusion of obstacle detection methods can effectively detect different types of obstacles and accurately measure and track their positions.