Vehicle Probe Data Analysis for Automatic Parking Location Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Mapping services face challenges in accurately identifying and updating vehicle parking, stopping, and standing locations on digital maps, which are typically resource-intensive and prone to inaccuracies due to manual data collection methods.
Innovation Solution
A system that uses probe data from vehicles to automatically detect and flag potential parking, stopping, and standing locations by analyzing vehicle behavior patterns, such as prolonged stationary periods and reoccurring vehicle appearances, and integrates this data into geographic databases using machine learning or naive algorithms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual data collection methods are used to identify parking locations, then map data can be updated, but the process becomes resource-intensive and time-consuming
Solution Approach 1:
The system enables self-service by automatically detecting parking locations through analysis of probe data from vehicles themselves. The probe data contains vehicle behavior information (speed, location, time) that directly indicates parking events, eliminating the need for external manual data collection. The system processes this self-generated data to automatically update map data with parking locations.
Solution Approach 2:
The patent replaces the mechanical manual data collection process with an automated electronic system. Instead of human collectors physically visiting locations to identify parking areas, the system uses electronic probe data from vehicles, processes it through automated algorithms, and generates map updates electronically, substituting the entire mechanical collection process.
2Reliability
If manual data collection methods are used to identify parking locations, then map data can be updated, but resource requirements increase significantly
Solution Approach 1:
The system leverages existing probe data that vehicles already generate and transmit during normal operation. This self-service approach means no additional resources are needed for data collection - the vehicles themselves provide the data as part of their normal functioning, and the system simply processes this readily available information to identify parking locations.
Solution Approach 2:
The system creates a digital copy of vehicle behavior patterns from probe data to identify parking locations. Instead of physically observing or manually recording parking events, the system analyzes copied data (speed, location, time information) from vehicle sensors to infer parking behavior, significantly reducing the resources needed compared to physical data collection methods.
3Productivity
If automated detection using probe data is implemented, then resource requirements are reduced, but measurement precision challenges arise in detecting parking events
Solution Approach 1:
The system uses parameter changes in probe data to detect parking events. Specifically, it monitors changes in vehicle speed (from moving to stationary), changes in location (vehicle remaining at same position), and temporal patterns (duration of stationary period). By detecting these parameter changes and applying threshold criteria, the system accurately identifies parking events from the continuous probe data stream.
Solution Approach 2:
The system employs feedback mechanisms to improve detection accuracy. It analyzes patterns in probe data over time, comparing current vehicle behavior against established criteria for parking events. The system refines its detection by considering multiple data points and temporal patterns, using feedback from the data analysis process to confirm or adjust parking event identification, thereby maintaining high precision in automated detection.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An approach is provided for automatic detection of vehicle parking, stopping, and/or standing locations (e.g., parking lots, alleys, loading zones, standing zones, etc.). The approach, for example, involves collecting probe data transmitted from a plurality of vehicles traveling on a road segment. The approach also involves processing the probe data to determine a vehicle behavior indicating a vehicle parking location, a vehicle stopping location, a vehicle standing location, or a combination thereof. The approach further involves flagging the road segment as being associated with the vehicle parking location, the vehicle stopping location, the vehicle standing location, or a combination thereof based on the vehicle behavior to update a geographic database.