3D Environment Object Filtering for Efficient AV Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The existing methods for generating synthetic three-dimensional (3D) environments for autonomous vehicle (AV) simulations are inefficient due to the computational cost of recreating every real-world object, which is not all influential to the AV's trajectory, and the time-consuming process of collecting and processing high-definition map sensor data.
Innovation Solution
A process that filters real-world sensor data to seed a synthetic environment generation, using influence scores to select only the most influential objects for inclusion, and utilizes real-world map tile data to generate synthetic 3D environments that include semantic labels, reducing computational expense while maintaining accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If every real-world object is recreated in the synthetic environment, then the simulation accuracy is improved, but the computational cost increases
Solution Approach 1:
The patent extracts only the influential objects from the complete real-world sensor data based on their influence scores. Objects are filtered and selected based on their relevance to the autonomous vehicle's trajectory, extracting only the necessary subset rather than recreating all objects. This reduces computational cost while maintaining simulation accuracy for critical objects.
Solution Approach 2:
The patent applies partial action by recreating only a portion of the real-world objects - specifically those with high influence scores - rather than all objects. The influence score threshold determines which objects are included, implementing a selective approach that balances accuracy with computational efficiency.
2Measurement precision
If high-definition map sensor data is collected and processed in detail, then the map accuracy is improved, but the data collection and processing time increases
Solution Approach 1:
The patent extracts essential information from sensor data during the synthetic environment generation process, rather than performing exhaustive processing of all sensor data. By using influence scores to identify important objects and using map tile data for basic scene structure, the system extracts only the necessary information, reducing processing time while maintaining accuracy for critical elements.
Solution Approach 2:
The patent performs preliminary filtering of sensor data to identify influential objects before generating the synthetic environment. By pre-computing influence scores and selecting objects in advance, the system reduces the amount of data that needs detailed processing during environment generation, thereby reducing overall processing time.
Data Source
AI summary
The disclosed technology provides solutions for generating synthetic 3D environments, In some aspects, the disclosed technology includes a process of collecting sensor data corresponding with a three-dimensional (3D) space; identifying objects in the sensor data; filtering objects for inclusion in the synthetic map based on (1) a corresponding influence score for each of the one or more of the objects and (2) a corresponding duration of presence of each of the objects in the sensor data; selecting one or more of the objects based on the filtering, wherein the corresponding influence score of each of the one or more of the objects is greater than a first threshold and the corresponding duration of presence of each of the one or more of the objects is greater than a second threshold; and generating the synthetic map including a 3D rendering of the one or more of the objects selected.