Robot Relocalization Using Visual-Laser Fusion in Repeated Scenes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot positioning technologies face challenges in unstructured and open scenes, such as repositioning in repeated scenes and low positioning accuracy, and require modifications to the environment for Outside-In methods, while Inside-Out methods struggle with robustness in complex environments.
Innovation Solution
A method involving data acquisition from a camera, inertial measurement, and laser radar to perform feature encoding, dimension reduction, and multi-sensor fusion for accurate robot repositioning, using image vector features and laser point cloud data to generate consistent pose initial values and matching results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If vision and laser SLAM positioning methods are used, then the robot can perform active positioning without environmental modification, but the positioning accuracy is insufficient in complex, unstructured, and repeated scenes
Solution Approach 1:
The patent merges visual positioning (using image data and feature encoding) with laser positioning (using point cloud data and ICP algorithm) to create a hybrid positioning system. The visual positioning provides adaptability to complex environments by recognizing features in unstructured scenes, while the laser positioning provides precise metric measurements. The two positioning results are fused through weighted fusion to achieve both high adaptability and high positioning accuracy simultaneously.
2Reliability
If traditional single-sensor positioning methods are used, then the system complexity is low, but the positioning robustness is insufficient in repeated and unstructured scenes
Solution Approach 1:
The patent introduces a data fusion module as an intermediary that processes and integrates data from multiple sensors (camera and laser radar). This mediator coordinates the information from different sensing modalities, resolves inconsistencies, and produces a unified positioning result. The fusion module manages the complexity of multiple sensors while enhancing positioning robustness through multi-source information integration.
Solution Approach 2:
The positioning system uses a composite approach by combining data from heterogeneous sensors (visual data and laser point cloud data) similar to how composite materials combine different substances to achieve superior properties. The visual-laser composite positioning system leverages the complementary strengths of each sensor type to achieve robust positioning in diverse environments while managing system complexity through structured data fusion.
Data Source
AI summary
A robot relocalization method and apparatus, and a storage medium and an electronic device. The method comprises: performing feature coding on local visual features of image data to obtain image vector features, and searching, in a pre-set key frame database, for the image vector features subjected to dimension reduction processing, so as to obtain a plurality of search results (S120); solving the pose of a robot according to each search result, so as to obtain a first pose initial value, and performing matching on laser point cloud data in a pre-set two-dimensional grid map according to speed data and displacement data, so as to obtain a first matching result (S130); and generating a second matching result according to the first pose initial value and the first matching result, and performing robot relocalization according to a second pose initial value, the first matching result and the second matching result.