Autonomous Driving Scene Reproduction Using Real and Virtual Obstacles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous driving systems face challenges in verifying the effectiveness of solutions to problematic scenes encountered on open roads due to the high randomness of these environments, making it difficult to reproduce and test the fixes on real vehicles.
Innovation Solution
A scene reproduction test method that involves obtaining problematic scene data, determining key and non-key obstacle information, generating a virtual obstacle model, and using real devices to simulate obstacles, allowing for a controlled reproduction of the scene to test an optimized autonomous driving system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous driving vehicles are tested on open roads, then real-world driving performance can be evaluated, but the randomness of open road scenes makes it difficult to reproduce problematic scenes for verification
Solution Approach 1:
The patent creates virtual copies of problematic scenes by extracting key elements (obstacles, road conditions, environmental factors) from real-world problematic scenarios and reconstructing them in a virtual test environment. This allows repeated reproduction of specific problematic scenes without relying on random encounters on open roads, thereby improving verification reliability while maintaining scene adaptability
Solution Approach 2:
The patent segments problematic scenes into discrete controllable elements including key obstacles, non-key obstacles, road conditions, and environmental parameters. By dividing the complex scene into manageable components, the system can independently adjust and reproduce specific problematic conditions, enabling reliable verification while adapting to different scene types
2Measurement precision
If all obstacles in problematic scenes are simulated using real devices, then scene reproduction accuracy is improved, but test system complexity and cost increase significantly
Solution Approach 1:
The patent applies different simulation approaches to different obstacles based on their importance: key obstacles that directly cause the problematic situation are reproduced using real devices for high accuracy, while non-key obstacles use virtual models. This selective approach maintains scene reproduction accuracy for critical elements while reducing overall system complexity and cost
3Reliability
If the autonomous driving system is optimized based on test results, then system performance is improved, but without scene reproduction, it cannot be verified whether the optimization solved the original problem
Solution Approach 1:
The patent performs preliminary extraction and preservation of problematic scene elements during initial testing, creating reusable virtual scene templates before optimization occurs. When verification is needed, these pre-prepared templates can be quickly instantiated and reproduced, enabling fast verification of optimization effectiveness without time-consuming scene reconstruction
Data Source
AI summary
A scene reproduction test method, apparatus, device and program product of an autonomous driving system are provided by the present application. The method includes: obtaining problematic scene data generated during a test on an autonomous driving vehicle equipped with a first autonomous driving system, and determining key obstacle information and non-key obstacle information according to the problematic scene data; generating, according to the non-key obstacle information, a virtual obstacle model, and determining, according to the key obstacle information, a real device used to simulate a key obstacle; where the real device is a device that actually exists in a test environment; performing, by utilizing the virtual obstacle model and the real device, a reproduction test on an autonomous driving vehicle equipped with a second autonomous driving system. In the solution, the test object is a real autonomous driving vehicle, and the key obstacle is a real device.