Scenario Description Language for Autonomous Vehicle Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Complex scenarios encountered by autonomous vehicles are difficult to replicate in real environments for testing, making it challenging to evaluate their performance effectively, especially when encountering unanticipated objects like balls, dogs, or people.
Innovation Solution
A domain-specific language (SDL) is developed to create simulated scenarios that can be executed by a computer system, allowing for the integration of simulated objects into real or simulated environments to test how autonomous vehicles respond to various conditions, merging sensor data with simulated data to assess performance without the need for physical testing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical testing is used to validate autonomous vehicle scenarios, then measurement precision and reliability are improved, but device complexity and difficulty of operation increase due to the need for complex real-world setup and safety concerns
Solution Approach 1:
The patent creates virtual copies of real-world scenarios through simulated environments that replicate physical conditions, objects, and situations. These digital twins allow validation of autonomous vehicle scenarios without physical testing, maintaining measurement precision while eliminating the complexity and safety concerns of real-world setup
Solution Approach 2:
The patent introduces a simulated environment as an intermediary between the autonomous vehicle software and the physical world. This intermediate layer allows scenario validation to occur in a controlled virtual space that mediates between software testing requirements and physical safety constraints
2Ease of operation
If simulated environments are used to test autonomous vehicles, then ease of operation and safety are improved, but measurement precision may deteriorate due to artificial conditions
Solution Approach 1:
The patent dynamically adjusts simulation parameters to match real-world conditions, including sensor noise characteristics, object physics properties, and environmental factors. By carefully controlling and varying these parameters, the simulation maintains measurement precision while preserving the ease of operation and safety benefits of virtual environments
Solution Approach 2:
The patent applies different levels of realism to different aspects of the simulation. Critical validation scenarios use high-fidelity models with accurate physics and sensor behavior, while less critical elements use simplified representations. This local differentiation maintains precision where needed while preserving overall simulation efficiency and ease of operation
3Productivity
If analytical solutions are used for simple scenarios, then productivity is improved, but adaptability deteriorates when facing complex scenarios that require simulation
Solution Approach 1:
The patent creates a dynamic validation system that automatically selects between analytical solutions and simulation based on scenario complexity. Simple scenarios are processed through fast analytical methods for high productivity, while complex scenarios automatically transition to simulated environments, providing both speed and adaptability across the full range of validation needs
Data Source
AI summary
A domain specific language for use in constructing simulations within real environments is described. In an example, a computing device associated with a vehicle can receive, from one or more sensors associated with the vehicle, sensor data associated with an environment within which the vehicle is positioned. In an example, the vehicle can be an autonomous vehicle. The computing device associated with the vehicle can receive simulated data associated with one or more primitives that are to be instantiated as a scenario in the environment. The computing device can merge the sensor data and the simulated data to generate aggregated data and determine a trajectory along which the vehicle is to drive based at least in part on the aggregated data. The computing device can determine instructions for executing the trajectory and can assess the performance of the vehicle based on how the vehicle responds to the scenario.