Autonomous Mobile Navigation Using SLAM and Positioning Tags
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous mobile apparatuses face challenges with inaccurate navigation in environments with high similarity, drastic changes, and lack of texture, leading to loss and drift issues, which can result in collisions and safety risks.
Innovation Solution
A control method that utilizes SLAM (Simultaneous Localization and Mapping) and positioning tags, such as one-dimensional or two-dimensional codes, to extract feature points, calculate three-dimensional camera and world coordinates, and generate map files to improve navigation accuracy, combining image data from cameras to optimize positioning and navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional navigation methods are used in environments with high similarity and lack of texture, then the navigation system is simple to implement, but the positioning accuracy deteriorates leading to loss and drift
Solution Approach 1:
The patent introduces positioning tags as intermediary objects in the environment to assist the autonomous mobile apparatus in positioning. These tags serve as mediators between the apparatus and the environment, providing reliable reference points for accurate localization even in textureless or highly similar environments where traditional visual odometry fails.
Solution Approach 2:
The patent changes the parameter of environmental features by introducing artificial positioning tags with distinct visual characteristics. These tags transform the environment from textureless and homogeneous to having distinct, detectable features, enabling accurate positioning without requiring complex environmental assumptions.
2Measurement precision
If SLAM is used for navigation in challenging environments, then positioning accuracy improves, but the system becomes more complex and computationally intensive
Solution Approach 1:
The patent merges SLAM technology with positioning tag recognition to create a hybrid navigation system. This combination leverages the global positioning capability of tags and the continuous localization strength of SLAM, achieving high accuracy while distributing computational load across multiple algorithms rather than relying on a single complex system.
Solution Approach 2:
The patent segments the navigation task into two parts: using positioning tags for coarse localization and SLAM for fine positioning and mapping. This segmentation allows each algorithm to operate in its optimal regime, reducing the computational burden on any single system component while maintaining high overall accuracy.
3Adaptability or versatility
If the autonomous mobile apparatus operates in environments with drastic changes and high similarity, then adaptability to different scenes is required, but traditional methods suffer from loss and drift
Solution Approach 1:
The patent makes the navigation system universal by introducing positioning tags that can be deployed in various environments (supermarkets, airports, computer rooms). These tags provide a common reference framework that works across different scene types, enabling the apparatus to adapt to diverse environments while maintaining reliable positioning through the same tag-based approach.
Data Source
AI summary
The present disclosure provides an autonomous mobile apparatus and a control method thereof. The method includes: starting a SLAM mode; obtaining first image data captured by a first camera; extracting a first tag image of positioning tag(s) from the first image data; calculating a three-dimensional camera coordinate of feature points of the positioning tag(s) in a first camera coordinate system of the first camera based on the first tag image; calculating a three-dimensional world coordinate of the feature points of the positioning tag(s) in a world coordinate system based on a first camera pose of the first camera when obtaining the first image data in the world coordinate system and the three-dimensional camera coordinate; and generating a map file based on the three-dimensional world coordinate of the feature points of the positioning tag(s).