高危场景下ORB-SLAM2稠密地图构建

Dense Map Construction in High Risk Scenarios Based on Improved ORB-SLAM2 Algorithm

针对经典视觉ORB-SLAM2稀疏建图存在精度低、实时性差等现实问题,提出了基于ORB-SLAM2算法架构的视觉定位与稠密建图算法,并通过部署在Jetson Xavier NX为核心板的嵌入式平台上进行建图精度与实时性验证,可满足高危场景下的清晰完整地图信息的构建需求,便于人们快速摸排风险点、制定救援方案和避免二次事故发生。

A visual localization and dense mapping algorithm based on the ORB-SLAM2 algorithm architecture is proposed to address the practical problems of low accuracy and poor real-time performance in sparse mapping of classic visual ORB-SLAM2.The algorithm is deployed on an embedded platform with Jetson Xavier NX as the core board for mapping accuracy and real-time verification,which can meet the construction requirements of clear and complete map information in high-risk scenarios that making it easy for people to quickly identify risk points Develop rescue plans and avoid secondary accidents.