Movable object and control method thereof
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Movable objects face challenges in accurately determining their location amidst changing environmental factors, such as variations in illumination or feature point extraction difficulties, using existing control methods.
Innovation Solution
A method and system for a movable object that involves acquiring omnidirectional images, converting them to panoramic images, and using signal strength to generate maps, which are then applied to algorithms like deep neural networks to determine location and orientation, enabling the object to adapt and navigate effectively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature point extraction is used for location determination, then location accuracy can be achieved, but environmental changes such as illumination variations cause recognition failures
Solution Approach 1:
The patent transforms the imaging parameter from conventional 2D plane images to 360-degree panoramic images, fundamentally changing the spatial representation parameter. This allows the system to capture complete environmental information around the movable object, making location determination robust against illumination changes and feature point variations in any single direction
Solution Approach 2:
The panoramic image serves multiple functions simultaneously: it provides complete environmental context for location determination, maintains accuracy under varying illumination conditions, and works across different environmental scenarios. The single panoramic capture replaces multiple conventional images that would be needed to achieve the same universality
2Device complexity
If conventional 2D images are used for mapping, then device complexity is reduced, but location recognition accuracy deteriorates under environmental changes
Solution Approach 1:
The patent applies spherical geometry by capturing 360-degree panoramic images that wrap around the movable object in all horizontal directions. This spherical field of view provides comprehensive environmental context, enabling accurate location determination even when specific features are obscured or illuminated differently, without requiring complex multi-camera arrays
3Measurement precision
If multiple separate control means are used, then control precision is improved, but ease of operation deteriorates
Solution Approach 1:
The patent merges multiple control functions (location determination, environmental mapping, navigation planning) into a single integrated control method that processes panoramic images and signal strengths unifiedly. This consolidation maintains precision by using comprehensive data while improving ease of operation by eliminating the need for separate control systems for each function
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed herein are a movable object and a movable object control method. The movable object control method may include acquiring an image of a movable object's surroundings, acquiring a signal having strength changing depending on a location of the movable object, generating a map on the basis of the signal and the image of the surroundings of the movable object, and applying the map to an algorithm to acquire a learned algorithm.