UGV Navigation Using Dual-Resolution Maps for Obstacle Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous navigation of unmanned ground vehicles (UGVs) faces challenges in efficiently planning and tracking paths while avoiding obstacles due to limited processing power and resource constraints, requiring a balance between map resolution and computational resources.
Innovation Solution
The use of two maps, a large map for path planning and a small high-resolution map for steering instructions, where cells are classified as traversable or non-traversable, with data crossing between maps to determine obstacle sizes and update the UGV's location relative to obstacles, leveraging a scanning device and Inertial Navigation System (INS) for accurate navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a high resolution map is used to represent the area, then the navigation accuracy and obstacle detection precision are improved, but the processing time and computational resources required increase significantly
Solution Approach 1:
The patent divides the navigation system into two segments: a large low-resolution map for path planning and a small high-resolution map for obstacle avoidance. This segmentation allows each map to be optimized for its specific purpose, reducing the overall computational burden while maintaining navigation accuracy.
Solution Approach 2:
The patent applies different resolution qualities to different parts of the navigation system. The large map uses lower resolution for efficient path planning, while the small map uses higher resolution for precise local obstacle detection and avoidance, optimizing resource allocation based on spatial requirements.
2Area of stationary object
If a large area is represented by the map, then the navigation coverage and destination reachability are improved, but the data storage and processing power requirements increase
Solution Approach 1:
The patent segments the map representation into a large low-resolution map for broad coverage and a small high-resolution map for detailed local information. This allows the system to cover large areas without storing excessive data, as the large map uses minimal resolution while the small map focuses computational resources on critical local zones.
Solution Approach 2:
The patent introduces a resolution dimension to the map representation, creating a multi-resolution system where the same physical area can be represented at different detail levels. This dimensional approach allows efficient storage by representing distant areas at lower resolution while maintaining high resolution only where needed.
3Manufacturing precision
If a high resolution map is used, then the path planning accuracy through narrow passages is improved, but the computational resources and processing time required increase
Solution Approach 1:
The patent applies high resolution processing locally only in the small map where the vehicle operates, while the large map uses lower resolution for overall path planning. This local quality approach ensures accurate path planning through narrow passages when needed, without requiring the entire navigation system to consume high processing power continuously.
Solution Approach 2:
The patent segments path planning into two stages: coarse path planning using the large low-resolution map, and fine-tuned local path adjustment using the small high-resolution map. This segmentation allows accurate navigation through narrow passages while distributing computational load efficiently across different processing stages.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The disclosed subject matter includes a method and system for navigating an unmanned ground vehicle (UGV), that include: generating, based on the scanning output data, a first map comprising a first group of cells and characterized by a first size; generating, based on the scanning output data, a second map representing an area smaller than that of the first map comprising a second group of cells, which are characterized by a second size being smaller than the first size; wherein each cell in the first group of cells and the second group of cells is classified to a class selected from at least two classes, comprising traversable and non-traversable, wherein the second part at least partly overlaps the first part; navigating the UGV based on data deduced from crossing between cells in the first map and second map.