Virtual Sensor Verification Using Real-Vehicle Data Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for verifying autonomous driving algorithms lack accuracy in simulating real sensor data, which is crucial for reliable operations of autonomous vehicles, as they often fail to accurately replicate real-world sensor inputs in virtual environments.
Innovation Solution
A method and system for verifying virtual sensor accuracy by synchronizing and comparing real sensor data from real vehicles with virtual sensor data generated in a simulated environment, using GNSS/INS data to reproduce real vehicles on a virtual road and synchronize sensor data across different equipment, allowing for precise comparison and validation of virtual sensor outputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If virtual sensor models are used in autonomous driving simulators, then verification speed and efficiency are improved, but accuracy of sensor data is deteriorated
Solution Approach 1:
The patent creates virtual copies of real sensor data by mapping real sensor outputs to virtual sensor models. The virtual sensor model is trained using real sensor data, and the accuracy is verified by comparing virtual sensor outputs with corresponding real sensor data from synchronized vehicle trajectories. This copying approach enables fast verification while maintaining accuracy through systematic validation.
Solution Approach 2:
The patent implements a feedback mechanism where virtual sensor data is continuously compared with real sensor data, and the virtual sensor model is adjusted based on the accuracy verification results. The system calculates accuracy metrics and uses this feedback to refine the virtual sensor model, ensuring that productivity gains do not compromise data accuracy.
2Measurement precision
If real sensor data is collected and synchronized with virtual environment data, then verification accuracy is improved, but system complexity and data processing requirements are increased
Solution Approach 1:
The patent introduces an intermediary synchronization system that uses GNSS/INS data as a common reference frame to align real and virtual sensor data. This intermediary approach simplifies the complexity by providing a unified time and space reference, making the synchronization process more manageable despite the increased data processing requirements.
Solution Approach 2:
The patent employs GNSS/INS data serving multiple functions: it synchronizes real sensor data, reproduces vehicle trajectories in the virtual environment, and provides a common reference frame for accuracy verification. This multi-functionality reduces overall system complexity by using a single data source for multiple verification purposes.
3Reliability
If virtual sensor models are verified using real vehicle data, then reliability of autonomous driving algorithms is improved, but time and resources required for verification are increased
Solution Approach 1:
The patent performs preliminary synchronization and calibration of virtual sensor models using real vehicle data before actual algorithm verification. By pre-establishing the accuracy of virtual sensors through comparison with real sensor data, the system ensures reliable verification results while reducing the time required during the actual algorithm testing phase.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
There is a method for verifying accuracy of a virtual sensor model for simulation based on reality information data. According to an embodiment, a virtual sensor verification method may acquire information on positions and states of real vehicles which are running on a real road, may acquire real sensor data generated in real sensors of a reality information acquisition vehicle from among the real vehicles, may reproduce the real vehicles on a virtual road as virtual vehicles, based on the acquired information on the positions and states, may acquire virtual sensor data outputted from virtual sensors mounted in a virtual information acquisition vehicle from among the virtual vehicles, and may verify the virtual sensors by comparing the acquired real sensor data and the virtual sensor data. Accordingly, accuracy of virtual sensor data which is supplied to a recognition, determination, control algorithm for autonomous driving may be measured and verified, so that accuracy on a result of verifying based on a simulator of an autonomous driving algorithm may be enhanced.