Moving Body Guidance Using Partial Target Image Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing unmanned aircraft landing systems, such as those using pattern recognition with light emitting devices, fail to guide aircraft accurately if only a portion of the target is captured, as they rely on the entire pattern for positional computation, leading to difficulties in landing in constrained areas like high-density residential zones.
Innovation Solution
A moving body guidance apparatus and method that detects either a target member image or feature member image from an image captured by a moving body, allowing for guidance control to move the body closer to the target or designated region, even if the entire target is not captured, by utilizing a detection unit and control unit to calculate and adjust the movement direction based on the captured image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all of the target pattern is used for landing guidance, then the positional relationship computation is accurate, but the system cannot function when the entire target is not captured in the image
Solution Approach 1:
The patent segments the target into multiple feature members (e.g., individual light emitting devices arranged in a pattern). The detection unit can identify and compute positional relationships using only a portion of these feature members rather than requiring the entire target pattern. This segmentation allows the system to function reliably even when only part of the target is captured in the image.
Solution Approach 2:
The patent implements partial action by enabling the guidance system to operate with a subset of the target's feature members. Instead of requiring complete capture of all light emitting devices, the system can compute positional relationships using only the visible portion of the target in the captured image, thus maintaining functionality under partial observation conditions.
2Measurement precision
If a complete target pattern is required for accurate positioning, then the measurement precision is high, but the system becomes unusable in constrained landing areas where the full target cannot be captured
Solution Approach 1:
By dividing the target into multiple independent feature members that can be individually detected, the system can compute positioning information using only the visible segments in the captured image. This enables operation in constrained areas where the complete target pattern cannot fit within the image frame.
Solution Approach 2:
The patent transitions from requiring two-dimensional complete pattern capture to enabling positioning through partial pattern recognition. The system can determine positional relationships by analyzing the spatial arrangement of visible feature members, effectively solving the positioning problem even when the full two-dimensional pattern is not observable.
3Reliability
If the system requires the entire target to be in view for guidance, then the positional computation is reliable, but the ease of operation is reduced in difficult-to-reach areas
Solution Approach 1:
The system performs guidance operations using only the portion of the target that is visible in the captured image. This partial action approach maintains guidance reliability while significantly improving ease of operation in constrained and difficult-to-reach landing areas, as the system no longer requires the operator to position the aircraft to capture the entire target.
Data Source
AI summary
Provided are a moving body guidance apparatus, a moving body guidance method and a computer-readable recording medium that are for accurately guiding a moving body to a target site. The moving body guidance apparatus has a detection unit 2 that detects, from an image 40 captured by an image capturing apparatus 23 mounted on a moving body 20, a target member image 42 captured of an entirety of a target member 30 or a feature member image 43 captured of an entirety or a portion of feature members 31 and 32 forming the target member 30, and a control unit 3 that performs guidance control for moving a set position 41 set with respect to the image 40 and indicating a position of the moving body 20 closer to the target member image 42 or closer to a designated region 44 set based on the feature member image 43.