Cooperative Visual SLAM Map Fusion Using Rendezvous and Non-Static Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collaborative SLAM systems face limitations in integrating information from multiple platforms without using visual markers or additional sensors, particularly in environments with varying camera viewpoints or similar scenes, and often discard non-static features that can provide useful information.
Innovation Solution
A collaborative visual SLAM system that utilizes rendezvous situations to identify platforms and fuse local maps into a global map using non-static features extracted from monocular cameras, employing an optimization algorithm to match and integrate platform data without markers or additional sensors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If visual markers or additional sensors (stereo cameras, RGB-D sensors) are used for rendezvous-based map fusion, then platform identification accuracy and map fusion reliability are improved, but device complexity and cost increase
Solution Approach 1:
The patent extracts and utilizes non-static features (features that change over time such as moving objects, changing lighting conditions, or temporary structures) from monocular camera images for platform identification and map fusion. This eliminates the need for additional specialized sensors while maintaining identification accuracy through clever use of available visual data
Solution Approach 2:
The monocular camera serves multiple functions: it captures images for single-platform SLAM, extracts non-static features for platform identification during rendezvous, and provides visual data for map fusion. This multi-functionality eliminates the need for specialized sensors while maintaining system reliability
2Device complexity
If non-static features are discarded in traditional SLAM systems, then processing complexity is reduced, but collaborative SLAM capability is worsened
Solution Approach 1:
The patent converts non-static features, which were traditionally considered harmful or problematic in SLAM systems due to their variability, into beneficial identification markers for platform recognition. By tracking how non-static features move and change across frames, the system identifies other platforms during rendezvous and uses this information for collaborative map fusion
3Device complexity
If multiple platforms operate independently without collaborative map fusion, then system complexity is reduced, but area coverage and mapping efficiency deteriorate
Solution Approach 1:
The patent introduces a ground station as an intermediary that receives data from multiple platforms, performs centralized map fusion using non-static features, and generates a unified global map. This intermediary coordinates the collaborative effort, enabling efficient large-area mapping while keeping individual platform complexity low
Data Source
AI summary
An image-based cooperative simultaneous localization and mapping system and method by using rendezvous. The image-based cooperative simultaneous localization and mapping method in a multi-platform system preforms: estimating a camera posture on the basis of an image input through a camera in a single SLAM scheme and manufacturing an local map by each of multiple platforms; extracting a non-static feature from the image and managing same; transmitting, to a ground station, the camera posture, the local map, and the non-static feature as platform data; determining, by the ground station, whether a rendezvous situation has occurred between one of the multiple platforms and the remaining platforms, on the basis of the platform data; and when the rendezvous situation occurs, fusing the local map received from two or more rendezvoused platforms into a global map.