Path Querying via Turn Probability and Segmented Road Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path querying systems face reliability issues due to outdated road weights in urban road networks, leading to suboptimal or undesirable query results.
Innovation Solution
A method that utilizes user historical trajectories to select path query results based on turn probabilities, eliminating the reliance on real-time road weights by dividing urban road networks into regions and calculating combined probabilities for road segment sequences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If path querying operations depend on road weights in urban road networks, then query results can be obtained based on current road network data, but the reliability of query results deteriorates when road weights are not updated in time due to changing factors such as road width and road surface quality
Solution Approach 1:
The system pre-calculates and stores turn probabilities for road segments based on historical trajectory data before querying is needed. This preliminary action allows the system to have reliable path recommendations ready without needing to update road weights in real-time, thus maintaining both efficiency and reliability.
Solution Approach 2:
The patent introduces turn probability as an intermediary metric that mediates between the road network structure and the path querying operation. Instead of directly using road weights that may be outdated, the system uses turn probabilities derived from historical data, which serve as a more reliable intermediary for making path decisions.
2Reliability
If user historical trajectories are utilized to calculate turn probabilities for path selection, then query result reliability is improved, but system complexity increases due to data processing requirements
Solution Approach 1:
The system segments the complex task of path recommendation into manageable components: (1) collecting historical trajectory data, (2) calculating turn probabilities for individual road segments, and (3) using these probabilities for path selection. This segmentation reduces overall system complexity by breaking down the data processing into independent, reusable modules.
Solution Approach 2:
The system performs preliminary processing of historical trajectory data to pre-calculate turn probabilities for all road segments in advance. This preliminary action transforms raw historical data into structured probability values that can be directly used during querying, significantly reducing the computational complexity during actual path recommendation operations.
3Productivity
If road network is divided into urban road network regions with designated spacing, then data processing efficiency is improved, but the complexity of network segmentation increases
Solution Approach 1:
The urban road network is segmented into discrete regions with designated spacing, creating a grid-like structure that simplifies data organization and processing. Each region can be independently processed, improving overall data handling efficiency while the regular spacing pattern keeps the segmentation logic relatively simple.
Solution Approach 2:
The system changes the parameter of spatial organization by introducing designated spacing between regions. This parameter change transforms the continuous road network into a discretized structure with regular intervals, making data processing more efficient while maintaining a simple, predictable segmentation pattern.
Data Source
Figure 1~2
Figure 3~4
AI summary
The present disclosure provides a path querying method and device, an apparatus and a non-volatile computer storage medium. In the embodiments of the present disclosure, query data comprising a departure location and a destination are obtained, and then M road segment sequences are obtained according to the query data so that according to the turn probability from each road segment in the at least one road segment included in said each road segment sequence to the road segment's neighboring road segment which is accessible to the destination, N road segment sequences can be selected from the M road segment sequences as the path query results. Since the query-querying operation is performed without depending on the road weights, this can avoid the problem in the prior art about unreasonable query results because the road weights of some roads cannot be updated in time, thereby improving the reliability of the path querying operation.