UAV Smart Landing Using Geometric and Semantic Ground Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in safely and autonomously landing when battery power is low or navigation systems fail, requiring a technique that avoids contact with people or property and selects a convenient landing spot.
Innovation Solution
The UAV utilizes onboard sensors to generate perception inputs for position estimation, incorporating geometric and semantic knowledge to choose a safe landing area, avoiding obstacles and ensuring a flat surface, with mechanisms for user-guided landing training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the UAV performs autonomous landing using onboard sensors and perception inputs, then the landing safety and autonomy are improved, but the complexity of the navigation system increases
Solution Approach 1:
The navigation system is divided into modular components: onboard sensors (cameras, LIDAR, GPS), perception input generators, position estimators, and landing controllers. Each module performs a specific function, allowing the system to achieve high reliability through specialized subsystems while managing overall complexity through clear separation of concerns.
Solution Approach 2:
The system performs preliminary position estimation and landing site selection before the actual landing occurs. The UAV continuously generates perception inputs and estimates its position during flight, preparing landing options in advance so that when landing is required, the decision can be made quickly and safely without last-minute complexity.
2Ease of operation
If the UAV selects a convenient landing spot using geometric and semantic knowledge, then the landing convenience and safety are improved, but the computational requirements and system complexity increase
Solution Approach 1:
The UAV applies different types of knowledge (geometric and semantic) to different aspects of landing site evaluation. Geometric knowledge assesses local physical characteristics like flatness and obstacle-free zones, while semantic knowledge evaluates contextual appropriateness such as permitted landing areas. This localized application of specialized knowledge improves landing convenience without requiring the entire system to handle all types of analysis uniformly.
Solution Approach 2:
The system introduces an intermediary evaluation layer that processes geometric and semantic knowledge separately before integrating them into a final landing site selection. This intermediary processing stage manages computational complexity by breaking down the decision-making into manageable steps rather than requiring simultaneous complex calculations of all factors.
3Measurement precision
If the UAV continuously monitors position and environment during flight, then the autonomous navigation accuracy is improved, but the energy consumption increases
Solution Approach 1:
The UAV performs position estimation and environmental monitoring at periodic intervals rather than continuously. Perception inputs are generated at discrete time steps, and position estimates are updated periodically, allowing the system to maintain adequate navigation accuracy while reducing energy consumption compared to continuous monitoring. The monitoring frequency is adjusted based on flight phase and criticality of navigation tasks.
Data Source
AI summary
A technique is introduced for autonomous landing by an aerial vehicle. In some embodiments, the introduced technique includes processing a sensor data such as images captured by onboard cameras to generate a ground map comprising multiple cells. A suitable footprint, comprising a subset of the multiple cells in the ground map that satisfy one or more landing criteria, is selected and control commands are generated to cause the aerial vehicle to autonomously land on an area corresponding to the footprint. In some embodiments, the introduced technique involves a geometric smart landing process to select a relatively flat area on the ground for landing. In some embodiments, the introduced technique involves a semantic smart landing process where semantic information regarding detected objects is incorporated into the ground map.