Semantic Configuration Spaces for Real-Time UAS Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned Aerial Systems (UAS) face challenges in efficiently determining clear and accessible landing and navigation spaces due to computationally expensive and data-intensive methods using 2D imagery and 3D models, which consume valuable flight time and resources.
Innovation Solution
The generation of semantic layers and configuration spaces for objects and navigation barriers, allowing for pre-computation of valid vehicle configurations to avoid collisions, enabling lightweight real-time navigation route planning without the need for object recognition from 3D point clouds or 2D images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If 2D imagery and 3D models are used to map clear spaces for UAS navigation, then navigation safety is improved, but computational cost and data processing requirements increase
Solution Approach 1:
The system pre-computes configuration spaces for each object type before UAS navigation, storing valid vehicle configurations in advance. This eliminates the need for real-time computation of 3D model collisions during flight, reducing computational burden while maintaining navigation safety through pre-validated configuration data.
Solution Approach 2:
The mapping space is segmented into multiple configuration spaces, each corresponding to a specific object type (buildings, trees, power lines, etc.). Each configuration space contains pre-computed valid vehicle configurations for that object type, allowing the system to query and combine only relevant segments rather than processing all 3D model data in real-time.
2Reliability
If 2D imagery and 3D models are used to map clear spaces for UAS navigation, then navigation safety is improved, but flight time is consumed by imaging and processing
Solution Approach 1:
Configuration spaces are pre-computed and stored during map creation, eliminating the need for the UAS to image and process 3D models during flight. The system only needs to query pre-computed configuration spaces and combine them in real-time, dramatically reducing flight time while maintaining navigation safety through pre-validated data.
Solution Approach 2:
Instead of using actual 3D model data during flight, the system uses simplified configuration space representations that capture essential collision avoidance information. These copies contain pre-processed validity data that enables quick queries without requiring the original complex 3D models or real-time imaging.
3Measurement precision
If detailed 3D point-clouds and mesh models are used for mapping, then accuracy of clear space identification is improved, but data processing requirements and computational resources increase
Solution Approach 1:
The system extracts only the essential collision avoidance information from detailed 3D models during map creation, storing it in simplified configuration spaces. Each configuration space contains pre-computed valid vehicle configurations that capture the necessary spatial constraints without retaining the full complexity of original 3D point-clouds and mesh models, reducing data processing requirements while maintaining identification accuracy.
Solution Approach 2:
The system transforms complex 3D spatial data into parameterized configuration spaces with discrete validity states. By changing the data representation from continuous 3D models to parameterized configuration sets, the system maintains the ability to accurately identify clear spaces while dramatically reducing the quantity of data that must be processed during UAS navigation.
Data Source
AI summary
Disclosed are systems and methods that generate individual semantic layers for different object types within an environment. For each object type semantic layer and for an aerial vehicle, a semantic configuration space may be formed that indicates all valid aerial vehicle configurations in which the aerial vehicle does not collide with objects of the object type that exists in the environment. Finally, a combined configuration space may be formed by overlaying multiple semantic configuration spaces for object types known to be within the environment. The resulting combined configuration space indicates a common area in which the aerial vehicle can navigate according to the configuration of the common area and not collide with any objects of the object types represented by the combined semantic configuration spaces.