UAV Landing Zone Point-Cloud Detection for Edge Gap Hazards
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous landing technologies for UAVs are insufficient in detecting dangerous regions like rooftops and cliffs during landing, leading to a risk of crashes due to inadequate detection of edge gaps in flat but hazardous areas.
Innovation Solution
An autonomous landing method and apparatus that utilize a point cloud distribution map obtained by a depth sensor to divide the landing region into designated areas, determining if the point cloud quantity in each area is below a threshold, and controlling the UAV to avoid or stop landing in hazardous regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If overall situation analysis of the planned landing region is performed, then the overall flatness of the landing region can be detected, but dangerous regions such as rooftops, cliffs and deep ditches with edge gaps cannot be detected
Solution Approach 1:
The patent divides the detection process into two stages: first performing overall situation analysis of the planned landing region to detect overall flatness, then performing local detection on specific regions of interest to identify dangerous areas with edge gaps. This segmentation allows the system to maintain both comprehensive coverage and detailed detection capability without overwhelming complexity
Solution Approach 2:
The patent transitions from two-dimensional image analysis to three-dimensional point cloud analysis by introducing depth information from depth sensors. This dimensional enhancement enables the detection of edge gaps and dangerous regions that are invisible in traditional 2D images, significantly improving detection accuracy for hazardous landing zones
2Measurement precision
If local detection is performed by dividing the detection region into designated regions, then detection accuracy is improved, but the detection process becomes more complex
Solution Approach 1:
The detection region is divided into multiple designated regions based on the overall situation analysis results. Each designated region is independently analyzed for local characteristics, enabling precise identification of dangerous areas while maintaining manageable processing complexity through region-based decomposition
Solution Approach 2:
The system performs preliminary overall situation analysis before conducting detailed local detection. This preliminary action identifies regions of interest that require further analysis, allowing the system to focus computational resources on critical areas and reduce overall processing complexity
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances detection accuracy, reducing the risk of crashes by identifying and avoiding dangerous regions during landing, thereby improving the safety of UAV landings.
Implementation Method 1
obtaining point cloud data of the planned landing region by using a depth sensor; and projecting the point cloud data to a two-dimensional plane, to obtain the point cloud distribution map
Data Source
AI summary
The autonomous landing method of a UAV includes: obtaining a point cloud distribution map of a planned landing region; determining a detection region in the point cloud distribution map according to an actual landing region of the UAV in the point cloud distribution map; dividing the detection region into at least two designated regions, each of the at least two designated regions corresponding to a part of the planned landing region; determining whether a quantity of point clouds in each of the at least two designated regions is less than a preset threshold; and if a quantity of point clouds in a designated region is less than the preset threshold, controlling the UAV to fly away from the designated region of which the quantity of point clouds is less than the preset threshold, or controlling the UAV to stop landing.