Maze Solving via Channel Boundary Graph Construction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional maze path solutions are inefficient and inaccurate, especially in complex mazes, due to high computational requirements and the lack of consideration for dead ends, leading to reduced solving efficiency.
Innovation Solution
A maze solving method based on line-surface spatial relations, involving raster-to-vector conversion, extraction of channel surface-shaped data, and construction of virtual connecting lines to identify alternative solution paths, which omits the computation-intensive midline extraction, allowing for efficient path finding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional path search methods are used for maze solving, then the method can handle general path search problems, but the solving efficiency and accuracy are not satisfactory, especially in complex mazes
Solution Approach 1:
The patent segments the maze into discrete channel regions and uses graph theory to represent these segments as nodes and edges. This segmentation allows the complex maze to be broken down into manageable components that can be systematically explored and evaluated, improving both efficiency and accuracy in finding optimal paths.
Solution Approach 2:
The patent transitions from traditional 2D pixel-based maze representation to a 1D channel-based representation by extracting the central line of each channel and representing it as a graph element. This dimensional transformation simplifies the problem while preserving essential path-finding information, enabling more efficient computation.
2Measurement precision
If traditional maze algorithms use maze path midline extraction, then the path can be identified, but a huge amount of computation is caused, greatly reducing maze solving efficiency
Solution Approach 1:
The patent extracts only the essential channel boundary information and represents it as a simplified graph structure, omitting the computationally intensive midline extraction process. By taking out only the necessary channel definitions and their topological relationships, the method achieves accurate path identification with significantly reduced computation time.
Solution Approach 2:
The patent creates a simplified copy of the maze structure using channel boundaries and graph representations rather than working with the original complex pixel data. This copied representation preserves the essential path-finding characteristics while being much easier and faster to process computationally.
3Ease of manufacture
If traditional path search methods are used without automatic filtering in dead end environments, then the method is simple to implement, but the solving accuracy is reduced due to inability to handle dead ends
Solution Approach 1:
The patent performs preliminary action by pre-processing the maze image to extract channel boundaries and construct the graph representation before path searching begins. This preliminary extraction identifies all channels and their relationships in advance, enabling the subsequent path search algorithm to automatically filter dead ends and identify optimal paths without complex real-time processing.
Data Source
AI summary
A maze-solving method includes converting extracted channel surface-shaped data into channel boundary lines; extending extension lines from two end points at a start point and at a terminal point, to two sides outside a maze, and constructing, outside the maze, a virtual connection line I and a virtual connection line II connecting base points on the extension lines of the start point and the terminal point; respectively enclosing a polygon I and a polygon II by means of the virtual connection line I and the channel boundary lines and by means of the virtual connection line II and the channel boundary lines, in which paths connecting the start point and the terminal point, between the polygon I and the polygon II are alternative solution paths for the maze; and selecting an alternative solution path with the shortest length as the optimal solution path for the maze.