Own-Position Error Estimation Using Map and Motion Error Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for estimating the position of a mobile object do not accurately consider the error in map data, leading to low accuracy in position estimation.
Innovation Solution
An own position error estimation device and method that includes an image acquisition part, an environmental map data creation part, and an own position estimation part, which calculates errors using first, second, and third error calculation parts to enhance accuracy by considering map data errors and triangulation-based environmental map data creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If map data error is not considered in position error calculation, then the calculation process is simple, but the accuracy of own position error is low
Solution Approach 1:
The error calculation process is divided into three distinct modules: first error calculation part (map data error), second error calculation part (imaging part position/posture error), and third error calculation part (combined total error). This segmentation allows each error source to be calculated and processed independently, making the complex calculation manageable while achieving high accuracy through comprehensive error consideration.
2Productivity
If environmental map data is created from multiple key frames captured at intervals, then the map data creation efficiency is improved, but the first error calculation requires additional processing
Solution Approach 1:
The first error information is calculated and stored in the environmental map data during the map creation process itself. This preliminary calculation allows the error data to be readily available when needed for own position error estimation, avoiding redundant calculations and reducing processing time during actual position estimation operations.
Data Source
AI summary
A position error estimation device including an image acquisition part acquiring an image of surroundings of a mobile object captured by an imager on the mobile object; an environmental map data creation part creating environmental map data from the image; and an own position estimation part calculating a current own position of the mobile object from the image captured after the environmental map data is created. The own position estimation part includes: a first error calculator calculating a first error of the own position from an error related to a map point calculated when the environmental map data is created; a second error calculator calculating a second error of the own position from the image captured while the mobile object is moving from positions and postures of the imaging part; and a third error calculator calculating a third error of the own position including the first and second errors.