UAV Multi-Pattern Landing Control for GPS-Free Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current unmanned aerial vehicle (UAV) landing systems lack precision, particularly in environments with limited visibility or complex terrain, and often rely on expensive equipment, such as GPS, which may not function effectively in all conditions.
Innovation Solution
A multi-pattern recognition system using a camera sensor to identify coded control information for precise landing, employing a combination of larger and smaller patterns positioned around the landing area, with the ability to adjust posture based on pattern size differences and perform low-altitude flights for recognition if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS or expensive equipment is used for landing, then landing precision is improved, but system cost increases
Solution Approach 1:
The patent creates a virtual copy of GPS functionality by using visual patterns on the ground that the UAV's camera can detect and decode. Instead of relying on expensive satellite navigation hardware, the system projects landing information through coded ground patterns that the camera captures and processes, achieving precise landing guidance without expensive equipment
Solution Approach 2:
The patent replaces expensive, complex landing systems with simple, inexpensive components: a camera sensor (already present in most modern UAVs), ground-based pattern markers, and image processing algorithms. This substitution uses low-cost objects to achieve the same functional outcome as expensive specialized equipment
2Device complexity
If single-pattern recognition is used, then system complexity is reduced, but landing reliability deteriorates in limited visibility or complex terrain
Solution Approach 1:
The patent divides the landing guidance function into multiple independent pattern elements (first pattern, second pattern, third pattern) with different characteristics. Each pattern can be independently recognized, and their combined information provides robust landing guidance that works in various visibility conditions and terrains, preventing single-point failure
Solution Approach 2:
The patent varies key parameters of the recognition patterns including size (large vs. small patterns), position (different locations around the landing area), and coding information (different data encoded in each pattern). This parameter diversity enables the system to maintain reliability across different environmental conditions by selecting appropriate patterns for each situation
3Reliability
If multi-pattern recognition system is implemented, then landing reliability is improved, but device complexity increases
Solution Approach 1:
The patent makes the camera sensor perform multiple functions: it captures images for navigation, detects landing patterns, decodes control information, and provides feedback for posture adjustment. By making the existing camera multi-functional rather than adding separate specialized sensors, the system achieves high reliability without proportionally increasing device complexity
Solution Approach 2:
The patent combines multiple pattern recognition functions into a unified image processing system. The same camera and processing pipeline handle detection of multiple patterns, decoding of various control information, and generation of landing guidance, reducing overall system complexity compared to having separate systems for each function
4Measurement precision
If precise landing control is achieved, then landing accuracy is improved, but adaptability to various environmental conditions deteriorates
Solution Approach 1:
The patent implements dynamic adaptability by allowing the UAV to adjust its recognition strategy based on environmental conditions. The system can switch between recognizing different pattern sizes, adjust recognition thresholds, and select appropriate patterns based on visibility and terrain, maintaining precise landing control across varying environments
Data Source
AI summary
Disclosed are a precise landing method using a multi-pattern in an unmanned aerial control system and an apparatus therefor. In an aspect of the present invention, a precise landing method using a multi-pattern of an unmanned aerial robot in an unmanned aerial control system includes receiving an image value from an outside and recognizing a multi-pattern in which control information for precise landing control has been coded based on the image value, obtaining control information when an ID value included in the control information indicates a landing point, moving to the landing point based on the control information, and performing landing at the landing point, and recognizing the multi-pattern again if the landing is not completed. A landing area for the landing of the unmanned aerial robot may include the landing point and the multi-pattern. The multi-pattern may include a first multi-pattern and a second multi-pattern. The first multi-pattern may have a greater size than the second multi-pattern.