Intersection Turn Cost Calculation for Accurate Arrival Time Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing navigation systems fail to accurately predict arrival times at a goal due to the lack of consideration for right and left turn costs, which vary significantly with real-time traffic events, leading to reduced accuracy in travel time calculations.
Innovation Solution
A travel time calculation server and system that receives real-time traffic data and stores intersection-specific travel time data associated with combinations of traffic events, allowing for the generation and use of accurate right and left turn costs based on current traffic conditions, including traffic jams, passage speeds, and other factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If VICS travel time data is used for route prediction, then basic travel time information is available, but right and left turn costs are not considered leading to reduced accuracy
Solution Approach 1:
The patent segments the travel time calculation into two distinct components: base travel time from VICS data and additional right/left turn costs. By separating these elements, the system can independently calculate and combine them, ensuring that turn-specific delays are not lost in the aggregate VICS data.
Solution Approach 2:
The patent performs preliminary calculation of right and left turn costs based on historical traffic patterns and intersection characteristics before the actual route prediction. This pre-computed turn cost information is then integrated with real-time VICS data to improve arrival time accuracy.
2Measurement precision
If right and left turn cost is calculated using historical traffic data, then some turn cost information is available, but it does not reflect real-time traffic events reducing precision
Solution Approach 1:
The patent implements a feedback mechanism where actual travel times from vehicles equipped with navigation systems are continuously collected and used to update the right and left turn cost databases. This real-time feedback loop ensures that the turn cost information reflects current traffic conditions while maintaining a historical record for pattern recognition.
Solution Approach 2:
The patent makes the turn cost calculation dynamic by adjusting weights between historical data and real-time probe data based on traffic condition variability. When traffic patterns are stable, historical data carries more weight; when conditions change rapidly, real-time data is emphasized, optimizing accuracy for varying traffic scenarios.
3Measurement precision
If more traffic event data is collected to improve turn cost calculation, then prediction accuracy improves, but system complexity increases
Solution Approach 1:
The patent extracts only the essential turn cost information from complex traffic data streams, focusing specifically on the additional time required for right and left turns at intersections. By isolating and storing only this critical parameter, the system avoids the complexity of processing and storing all possible traffic variables while maintaining prediction accuracy.
Solution Approach 2:
The patent transforms complex multi-dimensional traffic event data into a simplified parameter representation - the turn cost value - which captures the essential impact of traffic conditions on turn performance. This parameter transformation reduces data complexity while preserving the information needed for accurate route prediction.
Data Source
AI summary
A travel time calculation server is provided which comprises: a probe data reception part 31 configured to receive travel time data of an intersection node 6 through which a vehicle 3 passes as well as a traffic event of each of links 1-4 connecting to the intersection node; and a right and left turn cost generating part 33 configured to store a table, which stores the travel time data of the intersection node 6 associated with combinations of traffic events of the links 1-4, in storage means 35.