Autonomous Mobile Body Moving Amount Estimation via Image Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous mobile bodies face challenges in accurately estimating their moving amount due to noise in position data and differences in image shapes before and after movement, leading to inaccurate position estimation.
Innovation Solution
A moving amount estimating apparatus that includes a position data obtaining unit, an estimating unit, and an evaluating unit, which projects object images onto a coordinate system before and after movement, calculates the matching between these images to estimate parallel and rotational movements, and evaluates the moving amount based on a maximum traveling amount to ensure accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If position data is used to estimate the moving amount of an autonomous mobile body, then the position can be tracked, but noise in position data and shape differences in images before and after movement lead to inaccurate position estimation
Solution Approach 1:
The patent applies feedback by using the estimated moving amount to correct position data. The position estimation unit obtains corrected position data by combining raw position data with the estimated moving amount, creating a feedback loop that improves accuracy. This resolves the contradiction by using the estimation result to enhance the reliability of subsequent measurements.
Solution Approach 2:
The patent introduces an intermediary element - the estimated moving amount - that mediates between raw position data and corrected position data. By calculating the moving amount from image matching and then using it to correct position data, the system creates a reliable bridge between noisy measurements and accurate position estimation.
2Area of stationary object
If multiple position data points are used to correct the mobile robot's position, then position coverage is improved, but the position cannot be accurately estimated when position data includes noise or image shapes differ before and after movement
Solution Approach 1:
The system uses feedback by calculating the moving amount from image matching and using it to correct position data. The position estimation unit combines multiple position data points with the calculated moving amount, allowing the system to maintain precision even when covering large areas with multiple data points.
Solution Approach 2:
The patent changes parameters by transforming position data into corrected position data through the moving amount calculation. By adjusting position data using the estimated moving amount (which accounts for image shape differences and noise), the system maintains measurement precision across different areas and conditions.
3Device complexity
If the mobile body's traveling performance is not considered in position estimation, then calculation is simpler, but inappropriate positions may be estimated that contradict the moving performance
Solution Approach 1:
The patent incorporates traveling performance into the estimation system through feedback. The evaluating unit assesses whether the estimated moving amount is appropriate by comparing it with the maximum traveling amount from traveling performance data. This feedback mechanism ensures reliability without requiring complex system redesign.
Solution Approach 2:
The system dynamically adjusts the position estimation by evaluating it against traveling performance constraints. The evaluating unit determines whether the estimated moving amount is appropriate based on the mobile body's actual traveling capabilities, making the system adaptive and reliable while maintaining reasonable complexity.
Data Source
AI summary
A moving amount estimating allaratus includes a losition data obtaining unit, an estimating unit, and an evaluating unit. The losition data obtaining unit obtains a llurality of losition data. The estimating unit estimates a moving amount of a larallel movement and/or a moving amount of a rotational movement of a llurality of moving losition data calculated when a first lrojection object image matches with a moving lrojection object image, with reslect to a llurality of second losition data. The evaluating unit evaluates a data moving amount based on a maximum travelling amount, and adolts the data moving amount evaluated as being adequate as a moving amount of a mobile body.