Probabilistic Safe Landing Area Determination for Aircraft
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous landing zone detection methods for aircraft, such as OPVs and UAVs, face challenges in accurately determining safe landing areas due to unpredictable conditions, particularly when video camera-based vision systems fail to provide sufficient information with high measurement uncertainty.
Innovation Solution
A probabilistic safe landing area determination system using Bayesian inference and cellular/grid-based representations of environmental terrain, integrating higher-order feature measurements from sensors to create a probabilistic safe landing area map, which updates probabilities in real-time and prioritizes safe landing areas based on mission models and constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If video camera-based vision systems are used for landing zone detection, then the system can identify landing areas, but the measurement precision and reliability are insufficient due to high uncertainty in unpredictable conditions
Solution Approach 1:
The patent combines multiple sensing modalities (video cameras, LIDAR, barometers, GPS) into an integrated sensing system. This merging of sensors allows the system to overcome the limitations of individual sensors by fusing their data, thereby improving both measurement precision and reliability in determining safe landing areas.
Solution Approach 2:
The system transforms raw sensor data into probabilistic safety parameters through Bayesian inference. By changing the parameter representation from deterministic sensor readings to probabilistic safety scores, the system can quantify uncertainty and make more reliable landing area determinations despite unpredictable conditions.
2Reliability
If multiple sensor measurements are integrated to improve landing area accuracy, then the reliability increases, but the device complexity and computational requirements increase
Solution Approach 1:
The patent divides the terrain into discrete cells and processes sensor data on a per-cell basis. This segmentation allows the complex computational task of analyzing entire landing zones to be broken down into manageable cell-level evaluations, reducing overall system complexity while maintaining reliability through systematic probabilistic assessment of each cell.
Solution Approach 2:
The system introduces probabilistic safety parameters as intermediary representations between raw sensor data and landing area decisions. These intermediate probabilistic scores serve as a bridge that simplifies the integration of multiple sensor modalities by providing a unified framework for combining heterogeneous data sources.
3Adaptability or versatility
If real-time probabilistic mapping is performed to update safe landing areas dynamically, then the adaptability to changing conditions improves, but the processing time and computational load increase
Solution Approach 1:
The system implements dynamic probabilistic mapping where safety parameters are continuously updated as new sensor measurements become available. This dynamic approach allows the system to adapt to changing conditions in real-time, with the probabilistic framework naturally handling temporal updates without requiring complete reprocessing of all data.
Solution Approach 2:
The system performs preliminary probabilistic assessment of potential landing areas during the approach phase, before final landing decisions are required. This preliminary action allows time-consuming probabilistic calculations to be performed in advance, reducing the processing time needed for final landing area selection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
According to an aspect of the invention, a method (300) of probabilistic safe landing area determination for an aircraft (100) includes receiving 302) sensor data indicative of current conditions at potential landing areas for the aircraft (100). Feature extraction (306) on the sensor data is performed. A processing subsystem (204) of the aircraft (100) updates a probabilistic safe landing area map (602) based on comparing extracted features of the sensor data with a probabilistic safe landing area model (214). The probabilistic safe landing area model (214) defines probabilities that terrain features are suitable for safe landing of the aircraft (100). A list of ranked landing areas (312) is generated based on the probabilistic safe landing area map (602).