Real-Virtual Driving Verification for Urban Autonomous Scenarios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional autonomous driving verification systems face challenges in replicating real-world road conditions and scenarios, leading to limitations in testing and verifying the performance of autonomous vehicles due to the risk of accidents and the inability to simulate complex urban environments with unexpected situations.
Innovation Solution
An autonomous driving verification system that converges real and virtual information to create a simulated environment, incorporating real-world data from sensors and virtual information to control vehicle functions, allowing for the simulation of various realistic situations and scenarios, including infrastructure elements like signal controllers and pedestrian detectors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous driving vehicles are tested on real roads, then real driving performance can be verified, but the risk of causing accidents increases
Solution Approach 1:
The patent creates a virtual copy of the real road environment that replicates physical infrastructure (roads, intersections, traffic signals), dynamic elements (vehicles, pedestrians, cyclists), and environmental conditions (weather, lighting). This virtual replica allows comprehensive safety testing without exposing real vehicles to actual road risks, while maintaining ecological validity through faithful reproduction of real-world scenarios
2Object-affected harmful factors
If a simulated environment is used for testing, then safety is improved, but the ability to implement real environment conditions is limited
Solution Approach 1:
The patent introduces a V2X communication unit as an intermediary that bridges the virtual simulation environment and real-world infrastructure. This unit receives actual road environment information from real traffic signals, vehicles, and roadside devices, then integrates it into the virtual environment. This intermediary layer allows the simulation to maintain safety while adapting to real environmental conditions through authentic data input
3Object-affected harmful factors
If conventional simulation systems are used, then safety is maintained, but complex urban scenarios with infrastructure cannot be effectively simulated
Solution Approach 1:
The patent segments the complex urban environment into distinct functional modules: physical infrastructure elements (roads, intersections, traffic signals), dynamic traffic participants (vehicles, pedestrians, cyclists), environmental conditions (weather, lighting), and communication infrastructure (V2X devices). Each segment can be independently configured, simulated, and verified, allowing comprehensive complex scenario testing while maintaining system manageability and computational efficiency
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The present invention provides an autonomous driving verification system through real-virtual information convergence, including: a vehicle driving unit that is composed of a plurality of devices that perform steering, driving, deceleration, and braking functions; a simulation unit that generates virtual information including a real city road environment; and an autonomous driving unit that verifies autonomous driving by constructing a virtual environment in which real surrounding environment information acquired from a sensor and virtual information of the simulation unit are converged, and controls the vehicle driving unit to be actually driven within the virtual environment to verify autonomous driving, in which the virtual information of the simulation unit may include signal display information received from infrastructure including at least one of a signal controller, an unexpected situation detector, a pedestrian detector, and CCTV within an intersection installed on a city street.