Autonomous Mobile Localization Using Lighting-Matched Environment Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous mobile apparatuses face significant performance deterioration in estimating their location when the environment changes, particularly due to variations in lighting conditions, as existing techniques are not robust enough to adapt to different lighting environments.
Innovation Solution
The autonomous mobile apparatus selects a suitable environment map from stored maps based on current environment information, such as lighting conditions, using image data from an imager and feedback signals, to accurately estimate its location through monocular simultaneous localization and mapping (SLAM).
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single environment map is used for location estimation, then the device complexity is reduced, but the location estimation accuracy deteriorates when lighting conditions change
Solution Approach 1:
The patent divides the single environment map into multiple environment maps captured under different lighting conditions. Each environment map is tagged with lighting information (dark, dim, normal, bright), allowing the system to select the appropriate map based on current lighting conditions. This segmentation resolves the contradiction by maintaining simple device architecture while improving location estimation accuracy through conditional map selection.
Solution Approach 2:
The patent changes the parameter of lighting conditions to select different environment maps. By capturing environment maps under varying lighting parameters (dark, dim, normal, bright) and using lighting detection to select the matching map, the system adapts to environmental changes without increasing device complexity. This parameter-based selection resolves the contradiction between simplicity and accuracy.
2Adaptability or versatility
If environment maps are captured under different lighting conditions, then the adaptability to varying environments is improved, but the quantity of stored data increases
Solution Approach 1:
The patent implements dynamic environment map selection based on real-time lighting conditions rather than storing all possible environment variations. The system dynamically detects current lighting conditions and selects the pre-captured map that best matches, reducing stored data volume while maintaining high environmental adaptability. This dynamic approach resolves the contradiction between adaptability and data quantity.
Solution Approach 2:
The patent performs preliminary action by pre-capturing environment maps under representative lighting conditions (dark, dim, normal, bright) before actual operation. This allows the system to have ready-to-use maps for different conditions without storing continuous variations, reducing data volume while maintaining adaptability. The preliminary capture of key environmental states resolves the contradiction efficiently.
3Reliability
If location estimation is performed using landmarks, then the robustness to external disturbances is improved, but the device complexity increases due to additional sensing requirements
Solution Approach 1:
The patent makes the imager serve multiple functions: capturing environment maps for location estimation, detecting current lighting conditions, and identifying landmarks. By using a single imaging device for multiple purposes rather than adding separate sensors, the system achieves robust landmark-based estimation without significantly increasing device complexity. This multi-functionality resolves the contradiction between reliability and complexity.
Solution Approach 2:
The patent merges the environment map capture function, lighting detection function, and landmark identification function into a unified imaging-based system. By combining these functions that could require separate sensors into a single imager workflow, the system achieves robust landmark-based location estimation while minimizing additional hardware complexity. This merging approach resolves the contradiction effectively.
Data Source
AI summary
An autonomous mobile apparatus includes a memory and a processor. The processor is configured to acquire environment information that is information of a surrounding environment of the autonomous mobile apparatus, based on the acquired environment information, select, as an estimation environment map, an environment map that is suitable for the surrounding environment from among environment maps that are saved in the memory, and estimate a location of the autonomous mobile apparatus using the selected estimation environment map and an image of surroundings of the autonomous mobile apparatus that is captured by an imager.


