UAV Multi-Marker Tracking for GPS-Independent Autonomous Landing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing UAV landing technologies are inadequate for precise and reliable autonomous landing, particularly due to reliance on GPS signals that may be unavailable or imprecise, and vision-based systems that struggle with limited field of view and marker detection.
Innovation Solution
A computer-implemented method for an unmanned aerial vehicle (UAV) that detects a target marker using multiple images, determines spatial relationships, and controls the UAV to approach the marker while maintaining the marker within the camera's field of view, using processors to generate horizontal and vertical control commands based on image data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If GPS-based landing control is used, then landing automation is achieved, but landing precision deteriorates due to signal unavailability or imprecision
Solution Approach 1:
The patent introduces visual markers as intermediary objects placed on the ground, which serve as mediators between the UAV and the landing target. These markers provide visual reference points that the UAV's camera can detect and track, enabling precise positioning without relying on GPS signals. The markers act as a bridge that translates the landing control problem from GPS coordinate-based to visual feature-based tracking.
Solution Approach 2:
The patent replaces the GPS-based electronic navigation system with a vision-based optical tracking system. Instead of using satellite signals and electronic coordinate processing, the system uses the UAV's camera to capture visual images of ground markers, processes these images to determine marker position and orientation, and uses this visual information for landing control. This substitution of mechanical/electronic systems with optical systems resolves the precision issue.
2Measurement precision
If vision-based marker detection is used, then landing precision is improved, but reliability deteriorates due to limited field of view and marker detection failures
Solution Approach 1:
The patent divides the landing area into multiple zones with different types of markers (e.g., approach markers, target markers, orientation markers). This segmentation allows the UAV to use different markers for different phases of landing: approach markers for initial acquisition and coarse positioning, and target markers for fine positioning. This division reduces the risk of complete detection failure and improves overall system reliability.
Solution Approach 2:
The patent implements preliminary marker detection and tracking during the approach phase before final landing. The UAV detects and tracks markers in advance, continuously updating its position and orientation information. This preliminary action ensures that the marker detection system is already engaged and calibrated before critical landing moments, providing a buffer against detection failures and improving reliability.
3Difficulty of detecting and measuring
If the imaging device field of view is widened to detect markers, then marker detection capability is improved, but vertical positioning precision deteriorates
Solution Approach 1:
The patent implements dynamic adjustment of the imaging device's field of view based on the UAV's distance from the ground markers. During the approach phase at greater distances, the field of view is widened to detect and acquire markers. As the UAV approaches and vertical positioning becomes critical, the field of view is narrowed or adjusted to provide higher resolution imaging of the markers, thereby maintaining both detection capability and positioning precision through dynamic adaptation.
Data Source
AI summary
A computer-implemented method for controlling an unmanned aerial vehicle (UAV) includes identifying a set of target markers based on a plurality of images captured by an imaging device carried by the UAV. The set of target markers includes at least two or more types of target markers that are in close proximity to be detected within a same field of view of the imaging device. The method further includes determining a spatial relationship between the UAV and the set of target markers based at least in part on the plurality of images, and controlling the UAV to approach the set of target markers based at least in part on the spatial relationship while controlling the imaging device to track the set of target markers such that the set of target markers remains within the same field of view of the imaging device.