Scenario Space Monitoring for Self-Driving ODD Safety
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current self-driving car technologies face challenges in determining the safety of unknown scenario data in real-time, making it difficult to assess whether the environment is suitable for automatic driving.
Innovation Solution
An electronic device and method that uses an encoding module and decoding module trained with scenario data and a loss function to generate a scenario space, allowing for the determination of whether current scenario data belongs to an operational design domain, thereby assessing safety and providing recommended actions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current self-driving car technology is used to determine safety of unknown scenario data, then the system can operate with existing technology, but it cannot determine in real time whether unknown scenario data is safe
Solution Approach 1:
The patent applies preliminary action by pre-training the encoding module and decoding module using comprehensive scenario data before actual operation. The system performs offline training to establish the scenario space and safety determination capabilities in advance, so that during real-time operation, the pre-trained model can quickly assess unknown scenarios without requiring real-time retraining or extensive computation.
2Reliability
If real-time determination of unknown scenario data safety is implemented, then safety of self-driving car is improved, but computational complexity and data processing requirements increase
Solution Approach 1:
The patent applies segmentation by dividing the safety determination process into distinct functional modules: an encoding module that processes raw scenario data into encoded representations, a scenario space that stores safety boundaries and operational design domains, and a decoding module that interprets encoded data for safety decisions. This modular architecture allows each component to be optimized independently and processed in parallel, reducing overall computational complexity while maintaining real-time safety assessment capabilities.
Data Source
AI summary
An electronic device and a method for determining scenario data of a self-driving car are provided. The method includes: obtaining training scenario data by using scenario data, a loss function and a self-driving program module; training an encoding module and a decoding module by using the training scenario data, and generating a scenario space by using the trained encoding module; obtaining a monitoring module by using the scenario space; and executing the monitoring module to determine whether current scenario data belongs to an operational design domain by using the current scenario data and the trained encoding module.