Mixed Reality Simulation for ADAS Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtualized simulation methods for testing Advanced Driver Assistance Systems (ADAS) are inadequate as they require extensive construction of virtual elements and lack realism, leading to inefficient and unsafe testing of autonomous vehicle control systems.
Innovation Solution
A mixed reality simulation system that combines real-world sensor stream data with 3D models to generate a simulated vehicle environment, allowing for the testing of ADAS systems by synthesizing real-world active agents and environmental elements with their virtual counterparts, thereby reducing the need for extensive virtual element construction and enhancing realism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If virtualized simulation testing methodologies are used to test ADAS systems, then testing cost is reduced and collision risk is eliminated, but the simulation lacks sufficient realism and variation to yield accurate and reliable results
Solution Approach 1:
The patent uses recorded real-world sensor data to create accurate copies of actual driving environments. Instead of generating entirely synthetic virtual worlds, the system captures real sensor streams from test vehicles and uses them as the foundation for simulation, preserving the authenticity and variability of real driving conditions while eliminating physical risks.
Solution Approach 2:
The system performs preliminary data collection by recording sensor streams during real-world driving before the simulation phase. This preliminary action captures authentic environmental variations and driving scenarios that can then be reused across multiple simulation tests, ensuring both realism and efficiency.
2Reliability
If real-world testing methodologies are used to test ADAS systems, then realistic driving scenarios are captured, but testing becomes time consuming, expensive, and risks collisions
Solution Approach 1:
The patent creates virtual copies of real driving scenarios by recording sensor data during actual driving and replaying it in simulation. This allows the same authentic scenarios to be tested repeatedly without additional real-world testing time or risk, maintaining scenario authenticity while dramatically reducing testing duration.
Solution Approach 2:
The system enables continuous simulation testing using recorded data, allowing multiple test iterations to run simultaneously or in sequence without requiring repeated real-world data collection. This maintains the authenticity of scenarios while eliminating the time constraints of physical testing.
3Adaptability or versatility
If virtualized simulations are constructed with extensive virtual elements, then testing coverage is improved, but processing time and computational resources increase
Solution Approach 1:
Instead of constructing complex virtual environments from scratch, the system copies real sensor data directly into the simulation pipeline. This approach achieves comprehensive testing coverage by leveraging the diversity of real-world data while minimizing computational overhead, as the data is already in the required format and contains authentic variations.
Solution Approach 2:
The recorded sensor data serves multiple functions simultaneously: it provides realistic environmental conditions, validates sensor processing algorithms, and can be reused across different simulation configurations and test cases. This multi-functionality achieves comprehensive testing coverage without requiring separate virtual constructions for each test type.
Data Source
AI summary
The disclosure includes a system, method, and tangible memory for generating a simulation. The method may include receiving real-world sensor stream data that describes a real-world data stream that is recorded by onboard sensors in one or more test vehicles, wherein the real-world sensor stream data describes real-world active agents and real-world environmental elements. The method may further include generating three-dimensional (3D) models of active agents and 3D models of environmental elements in a vehicle environment. The method may further include generating a simulation of the vehicle environment that synthesizes the real-world active agents with 3D models of active agents and synthesizes the real-world environmental elements with the 3D models of environmental elements.