Simultaneous Localization and Mapping Using Point-Line Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing SLAM technologies face challenges in achieving real-time performance and accuracy, particularly in environments with poor textures, where feature-point extraction and matching are limited, leading to decreased tracking accuracy and increased drift errors.
Innovation Solution
A simultaneous localization and mapping device that captures color and depth images, estimates an initial pose using feature points and line segment features, constructs a three-dimensional map, and determines a final pose using a preset frame set and time domain window, incorporating a point and line optimization combination to create collinear constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If feature-point extraction and matching are used for tracking, then localization can be achieved, but tracking accuracy decreases and drift errors increase in environments with poor textures
Solution Approach 1:
The patent combines feature-point extraction with line segment feature extraction and matching. By merging these two feature extraction methods, the system can reliably track in environments with poor textures where traditional feature-point methods fail, thus improving tracking accuracy while overcoming the limitations of texture-poor environments
Solution Approach 2:
The patent creates a composite feature representation by combining point features and line segment features. This composite approach leverages the complementary strengths of both feature types, enabling robust tracking in diverse environments including those with poor textures, thereby resolving the contradiction between reliability and detection difficulty
2Measurement precision
If nonlinear optimization is used to improve localization accuracy, then better precision is achieved, but real-time performance deteriorates
Solution Approach 1:
The patent segments the optimization process by dividing the feature set into point features and line segment features, and potentially processing them in separate stages or with different optimization weights. This segmentation allows the system to maintain localization accuracy while reducing the overall computational burden to achieve real-time performance
Solution Approach 2:
The patent applies partial optimization by focusing computational resources on the most critical features or by performing optimization on a subset of features rather than all features simultaneously. This partial action approach maintains sufficient localization accuracy while ensuring real-time performance requirements are met
3Weight of moving object
If monocular cameras are used for SLAM, then the system remains lightweight, but scale estimation of motion trajectory becomes difficult
Solution Approach 1:
The patent introduces line segment features as an intermediary element that provides additional geometric constraints for scale estimation. These line segment features act as mediators between the monocular camera observations and the motion trajectory scale, enabling accurate scale estimation while maintaining the lightweight nature of monocular vision systems
Data Source
AI summary
A simultaneous localization and mapping device is provided. The device includes an image obtaining device configured to capture color images and depth images of a surrounding environment; an initial pose estimating device configured to estimate an initial pose based on the color images and the depth images; a map constructing device configured to construct a three-dimensional map based on the depth images and the color images; and a pose determining device configured to determine a final pose based on the initial pose and the three-dimensional map.