Vehicle Overspeed Prediction Using Historical Trajectory Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting overspeeding in vehicles are resource-intensive and require manual input of speed limits, which is inefficient, especially in areas without defined legal speed limits.
Innovation Solution
A method and system that use historical trajectory data from a database to determine speed distributions in geographical areas, allowing electronic devices associated with vehicles to detect if the current speed exceeds a threshold speed based on pre-determined percentiles, and calculate the probability of future overspeeding using a trained classifier.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual input methods are used to define speed limits in map data, then speed limit information can be provided for detection, but the process becomes resource-intensive and inefficient
Solution Approach 1:
The system enables automatic speed limit detection by having vehicles themselves collect and report speed data through their sensors and trajectory tracking, eliminating the need for manual speed limit input. The vehicles serve their own purpose of data collection while the system processes this self-generated data to determine speed limits automatically
Solution Approach 2:
The patent replaces the manual mechanical process of speed limit input with an automated electronic system that uses machine learning models and sensor data processing. The system substitutes human operators with algorithms that automatically detect speed limits by analyzing vehicle trajectory and speed information from the fleet
2Reliability
If existing techniques are used to enrich map data with speed limits, then speed detection can be performed, but significant computational resources and manual effort are required
Solution Approach 1:
The system extracts only the necessary speed limit information from the collected trajectory data using machine learning models, rather than processing and storing all raw data. The extraction process identifies and isolates speed limit values from the complex dataset, reducing the burden on system resources while maintaining reliability
Solution Approach 2:
The patent changes the parameters of the detection system by using relative speed thresholds based on percentile rankings of historical speed data, rather than relying on pre-defined absolute speed limit values. This parameter transformation allows the system to adapt to different geographical areas without requiring manual speed limit input for each location
3Measurement precision
If manual speed limit input is required for all geographical areas, then accurate speed detection can be achieved, but the process is inefficient especially in areas without defined legal speed limits
Solution Approach 1:
The system creates a universal speed detection method that works across all geographical areas regardless of whether defined speed limits exist. The machine learning model serves multiple functions: it detects speed limits in areas with legal limits, infers appropriate speed thresholds in areas without posted limits, and adapts to different road types and conditions, making the system universally applicable
Data Source
AI summary
A method of detecting overspeeding for a vehicle, including obtaining historical trajectory data of a fleet, of geographical areas from an electronic database; determining, by a microprocessor of a server, a distribution of speed of the historical trajectory data; calculating, on an electronic device associated with the vehicle, a determined probability of future overspeeding; wherein obtaining includes communicating, by the server an electronic request to the electronic database for the historical trajectory data, and the historical trajectory data from the electronic database to the server. An electronic device including a trajectory data acquisition circuit; a communication circuit to receive pre-trained weights for a trained classifier from a server; a processor to use a classifier configured with the pre-trained weights to calculate, based on trajectory data, a probability of future overspeeding being higher than a pre-determined threshold. A system and a computer-readable medium storing computer executable code for the method.