Citrus Positioning Method Based on Camera and Lidar Data Fusion

Chengjie Xu,Can Wang,Bin Kong,Bingliang Yi,Yue Li
DOI: https://doi.org/10.1007/978-981-19-2448-4_15
2022-01-01
Abstract:There have been some research achievements in the field of citrus picking robot research, but there are problems of inaccurate positioning in the application scenarios of citrus picking robots. To solve this problem, a YOLO V3 convolutional neural network is used to detect citrus and locate the pixel coordinates in this paper. On the basis of joint calibration of the camera and lidar, the image and point cloud data were fused. A method for obtaining the depth value of the citrus center based on the fusion of camera and lidar data was proposed. The proposed method is verified in the experimental scene, and the positioning errors do not exceed 0.01 m, which provides a technical method for the research and development of citrus picking robots.
What problem does this paper attempt to address?