Vision-Laser Fused 2.5D Map Building for SLAM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current SLAM technologies face challenges in industrial and dynamic environments due to inaccurate initial values, low precision wheel-type odometers, and the inability to fuse stable open-source 2D laser and visual information, leading to insufficient expression dimensions and instability in map building for autonomous robots.
Innovation Solution
A vision-and-laser-fused 2.5D map building method that calculates inter-image frame transformations using RGB-D images, performs scanning matching with initial estimations, detects loop closures, and updates visual and grid dimensions to create a comprehensive 2.5D map, avoiding single sensor failures and error accumulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laser SLAM is used in structured scenarios, then scanning matching can be performed, but wheel-type odometer has low precision and accumulated error
Solution Approach 1:
The patent combines laser SLAM and visual SLAM into a fused system where laser data provides accurate positioning information and visual data provides continuous motion estimation, compensating for the weaknesses of each individual sensor and resolving the precision-reliability contradiction
Solution Approach 2:
The patent introduces visual feature tracking as an intermediary to provide continuous pose estimation that bridges gaps in laser data and corrects accumulated errors from wheel odometers, serving as a mediator between the two sensing modalities
2Adaptability or versatility
If visual SLAM is used, then navigation function is obtained, but it cannot be put into practical use due to lack of stable fusion with laser data
Solution Approach 1:
The patent merges visual SLAM's navigation capabilities with laser SLAM's measurement accuracy by fusing their respective data streams and optimization results, creating a system that maintains practical usability while enhancing adaptability to various environments
3Device complexity
If single sensor is used, then system complexity is reduced, but sensor failure causes map building instability
Solution Approach 1:
The patent dynamically adjusts the weighting and contribution of each sensor's data based on their relative reliability and environmental conditions, allowing the system to maintain stability by compensating for sensor degradation or failure through parameter adjustment rather than structural complexity
Data Source
AI summary
A vision-and-laser-fused 2.5D map building method includes: calculating inter-image frame transformation according to an RGB-D image sequence to establish a visual front-end odometer; taking a visual front-end initial estimation as an initial value of scanning matching, and performing laser front-end coarse-grained and fine-grained searches; performing loop closure detection, and performing back-end global optimization on a 2.5D map according to a detected closed loop; and performing incremental update on visual feature dimensions of the 2.5D map, and performing occupation probability update on grid dimensions. The 2.5D map is built using a method of fusing a laser grid and visual features, and compared with a pure laser map and a pure visual map, richness of dimensions is improved, and completeness of information expression is improved; the 2.5D map building method is not influenced by single sensor failure, and can still stably work in a scenario of sensor degradation.