3D Road Map Visualization for Autonomous Driving Simulation Debugging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current virtual environments for autonomous driving require significant time and cost for data cleaning, mapping, and three-dimensional modeling, and developers face challenges in identifying errors in autonomous-driving simulations due to the cumbersome process of searching for source data and metadata.
Innovation Solution
A method and system for rendering a three-dimensional virtual environment based on high-definition road map data, visualizing metadata such as links, nodes, and traffic light information, and allowing users to interact with this data through user inputs to identify and resolve simulation errors efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If virtual environments are constructed manually with data cleaning, mapping, and three-dimensional modeling, then the accuracy and detail of the simulation environment is improved, but the time and cost required for construction increases significantly
Solution Approach 1:
The patent uses high-definition map data as a template to automatically generate virtual environment elements, copying real-world road geometry, traffic structures, and signage into the simulation environment without manual modeling. This copying approach maintains high accuracy while dramatically reducing construction time and cost.
Solution Approach 2:
The system performs preliminary processing of high-definition map data to extract and structure road information, traffic light positions, and other environmental elements before simulation runtime. This preliminary action prepares the environment in advance, eliminating the need for time-consuming manual construction during actual testing phases.
2Reliability
If developers manually search for source data and metadata to verify simulation errors, then comprehensive error verification is achieved, but the process becomes cumbersome and time-consuming
Solution Approach 1:
The patent introduces an intermediary system that automatically links simulation errors to their corresponding source data and metadata. When an error occurs, the system acts as a mediator to retrieve and display the relevant high-definition map data and metadata directly at the error location, eliminating the need for developers to manually search through multiple data sources.
Solution Approach 2:
The system provides immediate feedback by automatically displaying relevant source data and metadata when simulation errors are detected. This feedback mechanism allows developers to verify errors comprehensively without manual searching, as the system actively presents the necessary information for verification at the point of error detection.
3Loss of information
If high-definition road map data is visualized with multiple metadata elements (links, nodes, traffic lights), then the information completeness for error identification is improved, but the display complexity and data processing load increases
Solution Approach 1:
The patent segments the visualization of metadata by allowing selective display of different elements (links, nodes, traffic lights) based on user needs and error context. Rather than displaying all metadata simultaneously, the system divides the information into manageable segments that can be individually controlled and displayed only when relevant to the current error or analysis task.
Solution Approach 2:
The system implements partial visualization by displaying only the necessary portion of metadata relevant to the current simulation error or analysis focus. This partial action approach avoids the complexity of rendering and managing all possible metadata elements, while still providing sufficient information for effective error identification and debugging.
Data Source
AI summary
A method for displaying three-dimensional virtual environment information for autonomous driving simulation includes rendering a three-dimensional virtual environment for a specific region based on high-definition road map data representing road information of the specific region, and outputting the rendered three-dimensional virtual environment on a display of the user terminal, outputting, on the display, a vehicle traveling using an autonomous driving algorithm within the three-dimensional virtual environment, and visualizing and outputting, within the rendered three-dimensional virtual environment on the display, at least a portion of metadata included in the high-definition road map data.