3D SCENE RECONSTRUCTION AND PATH PLANNING METHOD FOR UAV IN GNSS-DENIED ENVIRONMENT

Today, natural disasters have a huge impact all over the world, while GNSS plays an important role in disaster relief and rescue. However, when the ground surface is severely damaged and covered, satellite positioning means are denied. In addition, disaster site conditions are often very complex and may require unmanned robots such as UAVs for pre-surveying. To address the raised problem, we reconstructed the 3D scene by laser SLAM; improved PRM path planning method for better computational efficiency while solving feasible path results; and realized UAV autonomous flight along the planned path in GNSS-denied environment. The experiments prove that the reconstructed scene map provides a feasible means for UAV autonomous navigation in GNSS-denied environment, and the proposed path planning method has a significant improvement in computational efficiency.