一种无人机动态语义地图构建方法、系统、设备及介质
By constructing a feature extraction and semantic segmentation model for UAV laser point cloud data and image data, and combining it with exponential decay weighted updates, the real-time and dynamic adaptability issues in the construction of dynamic semantic maps for UAVs were solved, achieving high-precision environmental perception and autonomous navigation support.
CN122415919APending Publication Date: 2026-07-17ELECTRIC POWER RES INST OF GUANGDONG POWER GRID CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- ELECTRIC POWER RES INST OF GUANGDONG POWER GRID CO LTD
- Filing Date
- 2026-04-20
- Publication Date
- 2026-07-17
Smart Images
Figure CN122415919A_ABST
Abstract
本发明公开了一种无人机动态语义地图构建方法、系统、设备及介质，属于智能体任务执行技术领域，所述方法为：对获取的激光点云数据进行特征提取和点云配准，通过估计优化不同时刻的无人机位姿，并将对应时刻的激光点云数据变换至全局坐标系，以构建全局点云地图；将获取的图像数据输入至预设的语义分割模型，提取所述图像数据的语义特征，输出语义分割图；将所述语义分割图投影至所述全局点云地图，获得语义三维点，并根据所述语义三维点构建初始语义地图；通过指数衰减加权进行动态更新，获得实时更新的目标语义地图，因此，通过实施本发明，能够实现高精度的环境感知与动态语义地图构建，为无人机自主导航提供实时准确的信息支撑。
Need to check novelty before this filing date? Find Prior Art