Tile-Based Correction Values for A* Route Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The A* algorithm used in vehicle navigation systems can take a long time to calculate the optimum route if the estimation function does not accurately represent real route and path conditions, leading to inefficiencies in route determination.
Innovation Solution
A method for determining correction values for route calculation algorithms based on digital map data, where the area is subdivided into tiles, and special nodes are predefined to calculate cost ratios and mean values, ensuring accurate and efficient route determination by adapting the heuristic values to comply with the A* algorithm's prerequisites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the estimation function does not accurately represent real route conditions, then the A* algorithm can calculate routes, but the calculation time becomes very long
Solution Approach 1:
The patent applies preliminary action by pre-calculating correction values for heuristic estimates before the actual route query is performed. These correction values are stored in the digital map data and automatically applied during route calculation to accelerate the A* algorithm without requiring real-time computation of accurate route conditions.
Solution Approach 2:
The patent changes the parameter of the heuristic estimation function by introducing correction values that adjust the estimated costs. These correction values modify the heuristic parameter to better reflect actual route conditions, improving both accuracy and calculation efficiency by reducing the number of nodes the A* algorithm needs to evaluate.
2Measurement precision
If correction values are calculated for all nodes in the area, then accuracy is improved, but the computational complexity and data storage requirements increase significantly
Solution Approach 1:
The patent applies segmentation by dividing the geographical area into multiple tiles, where correction values are calculated and stored only for boundary nodes of these tiles. This segmentation reduces the total number of nodes requiring correction values from all nodes in the area to only the boundary nodes, significantly reducing computational complexity and data storage requirements while maintaining sufficient accuracy for route calculations that cross tile boundaries.
Solution Approach 2:
The patent applies local quality by calculating correction values specifically for boundary nodes of tiles rather than uniformly for all nodes. This localized approach focuses computational resources on the most critical nodes where routes are most likely to transition between different heuristic estimation zones, improving efficiency without sacrificing overall route calculation accuracy.
Data Source
AI summary
A predefined area is divided into a plurality of tiles, a respective one of the tiles of the area being determined as a starting tile and the respective remaining tiles being determined as reference tiles, and special nodes, being predefined that are arranged in the starting tile or in the respective reference tile. For the respective starting tile, a correction value is determined, in each case in relation to one of the reference tiles, such that a respective cost ratio is determined for each special node of the respective starting tile in relation to each special node of the reference tile on the basis of the edge costs of the connections that are situated between the respective special node of the starting tile and the respective special node of the reference tile. The cost ratio is also determined on the basis of a heuristic value of the respective special node of the starting tile in relation to the respective special node of the reference tile. A mean value is determined on the basis of the determined cost ratios. The mean value is taken as a basis for determining the correction value of the starting tile in relation to the respective reference tile.