3D Distance Sensor Feature Map for Mobile Robot Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current feature map creation methods for mobile robots are sensitive to variations in the environment, such as image scaling, rotation, and affine curvature due to lighting and robot location changes, limiting their localization capabilities.
Innovation Solution
A feature map creation apparatus and method using a 3D distance sensor to detect distance and remission information, with corner, planar patch, and reflection function detection, allowing for the extraction and storage of feature information that is less sensitive to environmental changes, utilizing corner detection algorithms like SIFT and plane detection via least squares methods, and curve fitting for reflection functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image information from camera is used to extract features, then feature extraction is performed, but matching capability is limited due to sensitivity to image scaling, rotation and affine curvature from lighting and robot location variation
Solution Approach 1:
The patent changes the parameters used for feature extraction from image-based features to 3D geometric features (distance, corner, planar patch, reflection function) that are invariant to lighting and viewpoint changes. This resolves the contradiction by using parameters that maintain matching precision while being adaptable to environmental variations.
Solution Approach 2:
The patent substitutes the optical-based camera system with a 3D distance sensor system that measures geometric properties directly. This replacement eliminates the sensitivity to lighting conditions and image transformations, achieving both precise matching and environmental adaptability.
2Reliability
If grid map is used to reflect and store position and shape of surrounding environment, then localization is achieved, but memory requirement increases drastically as robot's use space becomes larger
Solution Approach 1:
The patent extracts only the essential geometric features (corners, planar patches, reflection functions) from the complete environmental representation, storing only these key features rather than the entire grid map. This reduces memory requirements while maintaining localization reliability through distinctive geometric landmarks.
Solution Approach 2:
The patent transitions from 2D grid map representation to 3D geometric feature representation, using depth information and spatial relationships to create a more compact and efficient map structure that requires less memory while providing robust localization.
3Quantity of substance
If feature map detects and stores only required feature information, then memory requirement is reduced, but system becomes relatively dependent on feature detection capabilities
Solution Approach 1:
The patent replaces complex image-based feature detection with simpler 3D geometric feature detection using distance sensors. The geometric features (corners, planes, reflection functions) are directly measurable from depth data, reducing detection difficulty while maintaining memory efficiency.
Solution Approach 2:
The patent changes the detection parameters from 2D image features to 3D geometric features that are inherently easier to detect and measure with distance sensors. This parameter change reduces the complexity of feature detection while preserving the compact representation benefits.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the robustness and efficiency of feature information extraction, enabling more accurate localization of mobile robots by reducing dependency on feature detection capabilities and improving matching capabilities across varying environmental conditions.
Implementation Method 1
a sensor to detect light reflected from surrounding objects to acquire distance information and remission information regarding the surrounding objects
Data Source
AI summary
Technology for creating a feature map for localizing a mobile robot and extracting feature information of surroundings is provided. According to one aspect, feature information including a reflection function is extracted from information acquired using a 3D distance sensor and used as a basis for creating a feature map. Thus, a feature map that is less sensitive to change in the surrounding environment can be created, and a success rate of feature matching can be increased.