Open-Door Asset Reconstruction for Realistic AV Sensor Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current simulation methods for testing autonomous vehicle control software lack realism, particularly when representing vehicles with open doors, leading to inadequate testing of autonomous vehicle interactions in dense urban environments.
Innovation Solution
The method involves determining the properties of an open door by fitting a rectangle to the door's polygon, adjusting the vehicle asset based on the door's absence, and positioning an open-door asset to create a more realistic simulation, allowing for better testing of autonomous vehicle control software.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current simulation methods are used to represent vehicles with open doors, then the simulation can be run quickly, but the realism of the simulation is insufficient
Solution Approach 1:
The vehicle model is segmented into multiple components: the main vehicle body and separate door assets. The door polygon is extracted and fitted with a rectangle to create an independent door asset that can be positioned and oriented separately from the vehicle body, enabling realistic representation of open door scenarios while maintaining manageable simulation complexity
Solution Approach 2:
Instead of using complex 3D models or detailed representations of open doors, the patent creates simplified rectangular assets that copy the essential geometric properties (position, orientation, dimensions) of the door polygons extracted from sensor data. This copying approach maintains realism while avoiding the complexity of full geometric reconstruction
2Measurement precision
If simplified representations of open doors are used in simulations, then the simulation complexity is reduced, but the testing accuracy for autonomous vehicle interactions is insufficient
Solution Approach 1:
The patent transforms the door polygon data into rectangular asset parameters (width, length, position, orientation) by fitting a rectangle to the polygon. This parameter transformation maintains the essential spatial characteristics needed for accurate testing while simplifying the representation to manageable complexity
Solution Approach 2:
The simulation applies different levels of detail to different parts of the scene: simplified rectangular assets for doors versus more detailed vehicle body models. This local differentiation allows testing accuracy where it matters most (door interactions) while keeping overall simulation complexity manageable
Data Source
AI summary
Aspects of the disclosure relate to generating assets for a simulation in order to test autonomous vehicle control software. For instance, data identifying a vehicle object, a polygon for the vehicle object, and a polygon for an open-door object associated with the vehicle object may be received. A rectangle may be fit to the polygon for the vehicle object. A polygon for the vehicle object without the open-door object may be determined based on the rectangle. A vehicle asset representing a vehicle may be adjusted based on the polygon for the vehicle object without the open-door object. A position and rotation of the rectangle may be used to position an open-door asset representing an open door for the simulation. The open-door asset is adjusted based on the adjustment to the vehicle asset. The simulation may be run with the positioned and adjusted open door asset in order to test the autonomous vehicle control software.