Autonomous Vehicle Simulation Using Virtual Perception Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current autonomous vehicle testing methods require significant resources and time, as they involve using real objects in a testing environment, which can be costly and inefficient, and do not allow for flexible or reproducible simulation of complex scenarios.

Innovation Solution

The use of simulated perception data to create a virtual environment for autonomous vehicles, allowing them to perceive and react to simulated objects, reducing the need for real-world objects and minimizing processing and memory requirements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If real objects are used in autonomous vehicle testing, then the testing realism and system reliability are improved, but the resource consumption, time requirements, and cost increase significantly

Engineering Contradiction:
Improvetesting realismVSAvoidtesting efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent creates virtual copies of real-world objects, environments, and scenarios through simulation. Instead of using physical objects for testing, the system generates digital representations that replicate real-world conditions, allowing autonomous vehicles to be tested in realistic scenarios without the resource-intensive requirements of physical testing setups.

Inventive Principle:
Principle #26Copying

2Reliability

If real objects and physical testing environments are used, then the authenticity of testing scenarios is improved, but the flexibility and reproducibility of complex scenarios deteriorate

Engineering Contradiction:
Improvescenario authenticityVSAvoidscenario flexibility
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The simulation system allows dynamic configuration and modification of testing scenarios without physical reconfiguration. Test environments, objects, and conditions can be changed programmatically, enabling flexible creation of complex scenarios while maintaining reproducibility through digital parameter adjustment rather than physical setup changes.

Inventive Principle:
Principle #15Dynamics

3Quantity of substance

If simulated perception data is used instead of real sensor data, then processing requirements and memory usage are reduced, but the complexity of generating realistic simulated data increases

Engineering Contradiction:
Improvememory requirementsVSAvoidsimulation system complexity
Core Design Contradiction:
Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent extracts only the essential perceptual features and data characteristics needed for autonomous vehicle testing from complex real-world scenarios. Rather than simulating every detail of physical environments, the system extracts key sensory information patterns that are sufficient for testing perception algorithms, reducing simulation complexity while maintaining testing effectiveness.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS10852721B1Autonomous vehicle hybrid simulation testing
Publication Date: 2020.12.01 AURORA OPERATIONS INC
  • US10852721B1 patent drawing
  • US10852721B1 patent drawing
  • US10852721B1 patent drawing

AI summary

Systems and methods for autonomous vehicle testing are provided. In one example embodiment, a computer implemented method includes obtaining, by a computing system including one or more computing devices, simulated perception data indicative of one or more simulated states of at least one simulated object within a surrounding environment of an autonomous vehicle. The computer-implemented method includes determining, by the computing system, a motion of the autonomous vehicle based at least in part on the simulated perception data. The computer-implemented method includes causing, by the computing system, the autonomous vehicle to travel in accordance with the determined motion of the autonomous vehicle through the surrounding environment of the autonomous vehicle.