Adaptive Hybrid Route Prediction Model for Trajectory Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing route prediction models fail to achieve satisfactory performance in various prediction situations due to their specific properties and limitations, such as reliance on history data, accuracy in routine trajectories, and inability to handle scenarios with no history data or insufficient data.
Innovation Solution
An adaptive hybrid model that combines object-specific, object group-specific, and object-independent prediction models, using decision rules and credibility parameters to determine the most accurate route prediction based on the current prediction context and route profile.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single prediction model (object-specific, object group-specific, or object-independent) is used, then the model has simple structure and easy operation, but the prediction accuracy is insufficient in various prediction situations
Solution Approach 1:
The patent combines multiple prediction models (object-specific, object group-specific, and object-independent models) into a hybrid prediction system. Each model type addresses different prediction scenarios, and their results are integrated through a decision rule to achieve higher overall prediction accuracy than any single model could provide alone.
Solution Approach 2:
The patent implements dynamic model selection by using a decision rule that adaptively chooses which prediction model to apply based on the current prediction situation and data availability. This dynamic approach allows the system to switch between different model types optimally rather than using a static single model.
2Measurement precision
If object-specific prediction models are used, then prediction accuracy improves for routine trajectories, but the system cannot handle scenarios with no history data or insufficient data
Solution Approach 1:
The patent segments the prediction problem into different model types based on data availability and trajectory characteristics. Object-specific models handle routine trajectories with sufficient history data, while object group-specific and object-independent models handle scenarios with limited or no history data, creating a segmented solution that addresses each scenario optimally.
Solution Approach 2:
The patent introduces object group-specific prediction models as an intermediary layer between object-specific and object-independent models. When an object lacks sufficient history data, the system uses group-level data as an intermediate resource to generate predictions, bridging the gap between individual and population-level analysis.
3Adaptability or versatility
If multiple prediction models are combined, then prediction accuracy and adaptability improve, but the system complexity and computational requirements increase
Solution Approach 1:
The patent applies local quality by using different model types in different prediction situations rather than uniformly applying a single complex system. The decision rule determines which model to use based on local conditions (data availability, trajectory type), simplifying the system's operation in each specific context while maintaining overall versatility.
Data Source
AI summary
A method, system, and computer program product for obtaining a first route traversed by a target object, performing at least one prediction for a second route to be traversed by the target object based on the first route, the at least one prediction being performed with at least one of an object-specific prediction model, an object group-specific prediction model, and an object-independent prediction model, and determining, according to a decision rule, a prediction result of the second route based on the at least one prediction.