Traffic Scenario Fusion for Automated Driving Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Testing and validating automated driving and driving assistance systems is highly complex and time-consuming, requiring prolonged road testing, and updates to these systems often necessitate revalidation, especially when considering different vehicle types.
Innovation Solution
Automatically generating a large number of simulation scenarios by extracting and augmenting recorded data streams corresponding to objects like vehicles, pedestrians, and obstacles, allowing for the simulation and validation of automated driving systems before failures occur in real-world traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If prolonged road testing is conducted to validate automated driving systems, then validation reliability is improved, but validation time and complexity increase significantly
Solution Approach 1:
The patent applies preliminary action by generating and executing simulation scenarios before actual road testing. The system creates virtual test environments with various traffic situations, obstacles, and edge cases that can be tested in advance, allowing validation of automated driving systems under controlled conditions before deploying to real-world roads, thus reducing the time and complexity of actual road testing while maintaining validation reliability
Solution Approach 2:
The patent uses copying by creating virtual replicas of real-world driving scenarios through simulation. Instead of testing every possible situation on actual roads, the system copies essential driving conditions, traffic patterns, and environmental factors into a virtual environment, allowing comprehensive validation without the time and resource constraints of physical road testing
2Adaptability or versatility
If extensive road testing is performed for different vehicle types, then system adaptability is improved, but testing complexity and resources increase
Solution Approach 1:
The patent applies universality by creating a multi-functional simulation platform that can test automated driving systems across different vehicle types using a single unified environment. The simulation system can configure various vehicle parameters, sensor layouts, and performance characteristics programmatically, allowing one testing infrastructure to validate multiple vehicle types without requiring separate physical test fleets for each vehicle model
Solution Approach 2:
The patent uses parameter changes by allowing dynamic modification of vehicle characteristics, environmental conditions, and scenario parameters within the simulation. Instead of physically reconfiguring test setups for different vehicle types, the system changes software-defined parameters such as vehicle dimensions, sensor positions, and performance metrics, enabling efficient adaptation testing across diverse vehicle platforms
3Reliability
If system updates require revalidation through road testing, then system reliability is maintained, but development productivity decreases
Solution Approach 1:
The patent applies preliminary action by implementing automated simulation-based regression testing that executes immediately after system updates. The simulation framework automatically re-runs relevant test scenarios with updated software versions, providing rapid validation feedback before deployment, thus maintaining system reliability while enabling faster update cycles compared to traditional road testing
Data Source
AI summary
Automated driving systems intended for use in a vehicle can be simulated and validated using automatically generated simulation scenarios. A large number of simulation scenarios can be generated by augmenting a plurality of recorded scenarios with one or more extracted data streams. The extracted data streams can correspond to objects such as other vehicles, pedestrians, cyclists, or other obstacles or barriers. The extracted data streams can be generated automatically by identifying similar recorded scenarios and isolating differences between the similar recorded scenarios.