Abstract
Compared with vision sensors, Lidar has a stronger anti-interference ability and higher measurement accuracy. However, when the multi-threaded lidar is far away from the obstacle, the point cloud data is sparse, it is difficult to locate effectively, and the obstacle cannot be classified. Because of the above problems, combined with the obstacle identification and positioning requirements when the live working robot of 500kV transmission line runs on the four-split conductor, we propose an obstacle recognition and localization method based on the fusion of Lidar and vision. This study uses the EfficientNet target detection algorithm to identify and classify obstacles and gets the area where the obstacle is located in the pixel coordinate system. The laser point cloud is projected to the pixel plane through the joint calibration of Lidar and camera. We extract and process the point cloud information in the target detection frame, then achieve accurate positioning of obstacles and provide reliable obstacle location information for live working robot to overcome obstacles autonomously. Experiments show that this identification and positioning method can effectively identify and classify obstacles, and has high positioning accuracy, which can meet the needs of live working robots to overcome obstacles autonomously.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
