Vision-Based Driving Scenario Generator for Autonomous Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving simulations are limited by the need for extensive manual configuration and the difficulty in capturing diverse real-life driving scenarios, which can be costly and time-consuming, especially when relying on data from a fleet of vehicles.
Innovation Solution
A system that generates driving scenario data using image data from cameras mounted on vehicles, analyzing two-dimensional image data to identify objects and determine their movement characteristics, which are then synchronized with GPS and inertial information to create three-dimensional driving simulations, allowing for extensive and varied scenario coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If manual configuration is used to create driving simulations, then the simulation can be developed with controlled scenarios, but the coverage of potential driving scenarios is limited and configuration effort is extensive
Solution Approach 1:
The patent uses real-world camera images as copies of actual driving scenarios to generate simulation data. Instead of manually creating hypothetical scenarios, the system captures and processes real visual data from vehicles, automatically converting it into simulation-ready format. This copying approach eliminates manual configuration while preserving the diversity and authenticity of real driving conditions.
Solution Approach 2:
The patent replaces the mechanical process of manual scenario configuration with an automated image processing system. Cameras mounted on vehicles capture driving scenarios, and computer vision algorithms automatically extract and structure the data, substituting human labor with optical and computational systems.
2Adaptability or versatility
If a fleet of autonomous driving vehicles is used to capture driving scenarios, then comprehensive real-life driving scenarios can be gathered, but the cost and complexity of data collection increase significantly
Solution Approach 1:
The patent makes camera systems universal by mounting them on regular vehicles that already exist in the fleet, rather than requiring specialized autonomous vehicles. The same camera hardware serves multiple purposes: capturing data for simulation generation, validating autonomous driving algorithms, and documenting real-world operating conditions. This multi-functionality reduces device complexity while maintaining scenario coverage.
Solution Approach 2:
The system enables vehicles to self-capture their own driving scenario data through onboard cameras. Each vehicle independently records and transmits its own operational data, eliminating the need for external observation vehicles or complex coordinated fleet operations. The vehicles serve themselves as data collection platforms.
3Measurement precision
If expensive equipment like LIDAR is used for data collection, then high-precision spatial data can be obtained, but the cost of the system increases significantly
Solution Approach 1:
The patent replaces expensive, durable equipment like LIDAR with inexpensive camera systems. While cameras have different measurement capabilities, they provide sufficient precision for simulation generation at a fraction of the cost. The system accepts that cameras capture two-dimensional images rather than three-dimensional point clouds, but this trade-off is acceptable given the cost reduction and the fact that simulations can be generated from 2D image data.
Data Source
AI summary
Described is a system (and method) for generating a driving scenario for an autonomous driving simulator. The system may use a camera mounted to a vehicle as a cost effective approach to obtain real-life driving scenario data. The system may then analyze the two-dimensional image data to create a three-dimensional driving simulation. The analysis may include detecting objects (e.g. vehicles, pedestrians, etc.) within the two-dimensional image data and determining movements of the object based on a position, trajectory, and velocity of the object. The determined information of the object may then be projected onto a map that may be used for generating the three-dimensional driving simulation. The use of cost-effective cameras provides the ability to obtain vast amounts of driving image data that may be used to provide an extensive coverage of the potential types of driving scenarios an autonomous vehicle may encounter.