Autonomous Mobile Self-Location Fusion for Changing Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-location estimation techniques for autonomous mobile bodies fail to accurately estimate location when the body frequently pauses or the external environment undergoes significant changes, such as day and night variations.
Innovation Solution
A self-location estimation device comprising a first unit that estimates the current self-location based on environmental map information and current image information, a second unit that estimates the self-location using learned parameters, and an integration unit that combines these estimates to provide a stable self-location even in changing environments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If landmark collation is used for self-location estimation, then location accuracy is improved under stable conditions, but the system fails when environmental appearance changes significantly
Solution Approach 1:
The self-location estimation system is divided into two independent modules: a first estimation unit using landmark collation and a second estimation unit using deep learning. Each module operates independently to estimate self-location, allowing the system to leverage the strengths of both approaches without mutual interference.
Solution Approach 2:
The patent merges the results from the first self-location estimation unit (landmark-based) and the second self-location estimation unit (deep learning-based) through integration. This combination allows the system to maintain accurate self-location estimation even when environmental conditions cause landmark appearances to change, as the deep learning component can compensate for failures in landmark collation.
2Device complexity
If a single estimation method is used, then system complexity is reduced, but estimation reliability fails under varying operational conditions
Solution Approach 1:
The system dynamically adapts to different operational conditions by employing multiple estimation methods that can be independently activated. The first estimation unit handles normal conditions with stable landmarks, while the second unit takes over when environmental changes affect landmark recognition, ensuring continuous reliable operation without requiring complex manual switching.
Solution Approach 2:
The patent changes the estimation approach parameters by switching between landmark collation-based estimation and deep learning-based estimation. This parameter change allows the system to adapt to varying environmental conditions, maintaining reliability whether the autonomous mobile body is moving continuously or frequently pausing, and regardless of environmental stability.
Data Source
AI summary
A self-location estimation device includes a first self-location estimation unit, a second self-location estimation unit, and a first integration unit. The first self-location estimation unit estimates a current first self-location of an autonomous mobile body based on current image information acquired by an image sensor and environmental map information stored in an environmental map information storage unit. The second self-location estimation unit estimates a current second self-location of the autonomous mobile body based on the current image information and a learned parameter learned using the environmental map information. The first integration unit estimates a current self-location of the autonomous mobile body by integrating the first self-location and the second self-location.