Mobile Body Pose Estimation Using Motion-Based Appearance Completion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the position and orientation of a mobile body via unmanned driving may fail to provide accurate results due to deficiencies in measurement information, such as blind spots caused by obstacles like workers, leading to incorrect estimations.
Innovation Solution
A device and method that includes an appearance information acquisition unit, movement information acquisition unit, determination unit, complement unit, and estimation unit to detect deficiencies in measured appearance information and use movement information to complement these deficiencies, enabling accurate estimation of position and orientation by comparing with reference information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If measurement information is directly used for estimation processing, then processing speed is maintained, but estimation accuracy deteriorates when deficiencies occur in measurement information
Solution Approach 1:
The system performs preliminary determination of whether deficiencies exist in measurement information before executing estimation processing. The determination unit assesses the quality of measurement information in advance, and only proceeds with estimation when the information is sufficient, preventing wasted processing on inadequate data
Solution Approach 2:
The determination unit acts as an intermediary between the measurement information acquisition and estimation processing. It evaluates whether the measurement information meets the required quality standards and controls the flow of data to the estimation unit, ensuring that only adequate information is processed
2Loss of information
If multiple detectors are added to eliminate blind spots, then measurement completeness improves, but device complexity and cost increase
Solution Approach 1:
The system uses its own movement information (from sensors already present for navigation) to self-correct deficiencies in measurement information. By combining movement data with partial appearance information, the system compensates for blind spots without requiring additional detectors
Solution Approach 2:
The system changes the parameter set used for estimation by incorporating movement information (position, orientation, velocity) as additional parameters to complement insufficient appearance information. This parameter substitution allows accurate estimation without adding physical detectors
Data Source
AI summary
A device includes: an appearance information acquisition unit configured to acquire measured appearance information on an appearance of a mobile body; a movement information acquisition unit configured to acquire movement information on movement of the mobile body; a determination unit configured to determine whether a deficiency has occurred in the measured appearance information; a complement unit configured to, in a case where the determination unit determines that the deficiency has occurred, use the movement information to complement the deficiency in the measured appearance information; and an estimation unit configured to estimate at least one of a position and an orientation of the mobile body by comparing reference appearance information on the appearance of the mobile body with the measured appearance information, and, in a case where the determination unit determines that the deficiency has occurred, compare the complemented measured appearance information with the reference appearance information.