Vehicle Distance Estimation During Sensor Gaps Using Historical Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining vehicle distance are limited by the spatial constraints of vehicle sensors, leading to inaccurate traffic density calculations and inadequate anonymization due to low traffic density, which compromises the quality of data used for automated driving functions.
Innovation Solution
A method that calculates average vehicle distance for a current observation period by extrapolating from a previous observation period using statistical assumptions, scaling the previous average distance proportionally to the duration ratio, even when sensor data is unavailable, ensuring a non-zero traffic density estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vehicle sensors are used to determine traffic density, then real-time data collection is achieved, but spatial constraints lead to inaccurate traffic density calculations
Solution Approach 1:
The system performs preliminary actions by storing historical traffic density data and average vehicle distance information from previous observation periods. When current sensor data becomes unavailable or insufficient, these pre-stored values are immediately utilized to maintain continuous traffic density estimation, thereby resolving the contradiction between real-time measurement and spatial constraints.
Solution Approach 2:
The patent introduces statistical assumptions and calculation models as intermediaries between sensor data and traffic density results. By using average vehicle distance from previous periods as an intermediary parameter, the system can estimate current traffic density even when direct sensor measurement is limited by spatial constraints, thus improving measurement precision beyond what the sensor's detection range alone would allow.
2Reliability
If data anonymization is performed with low traffic density estimation, then data protection is achieved, but data quality and functionality for automated driving validation deteriorate
Solution Approach 1:
The system dynamically changes the parameter of traffic density estimation by combining real-time sensor data with historical average vehicle distance information. This parameter optimization ensures that traffic density values remain non-zero and realistic even during periods of limited sensor detection, thereby maintaining data quality for automated driving validation while still enabling effective anonymization through statistical representation.
3Reliability
If extrapolation from previous observation period is used, then traffic density estimation is maintained during sensor unavailability, but accuracy may deteriorate without proper scaling
Solution Approach 1:
The patent applies dynamics by making the extrapolation process adaptive rather than static. The system dynamically scales the average vehicle distance from previous periods based on the ratio of current to previous observation period durations. This dynamic adjustment ensures that extrapolated values remain proportionally accurate and reflect current traffic conditions, preventing degradation of measurement precision while maintaining estimation continuity.
Solution Approach 2:
The system incorporates feedback mechanisms by continuously monitoring the availability of current sensor data and comparing it with historical patterns. When sensor data becomes unavailable, the feedback loop triggers the extrapolation process using previously stored average vehicle distance information, and when data becomes available again, the system validates and updates its estimates, thereby maintaining both reliability and precision through continuous adaptation.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present invention relates to a method, a computer program with instructions, and a device for determining a vehicle distance for an observation period. The invention further relates to a motor vehicle and a backend in which a method or device according to the invention is used. In a first step, measured values for an average vehicle distance are received for a plurality of measurement times (10). These can be stored for later use (11). It may occur that an unusable measured value is detected at an error time (12). In this case, a vehicle distance for a current observation period, which includes the error time, is determined based on a vehicle distance calculated from the recorded measured values for a previous observation period (13).