Autonomous Vehicle Simulation for Driver Behavior Gap Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle control software lacks effective methods for testing and validation to ensure safe and efficient interaction with other road users, leading to potential differences in driving behavior compared to manually-driven vehicles.
Innovation Solution
A simulation method is employed to compare the progress of a manually-driven vehicle with a simulated autonomous vehicle by generating path segments from log data, extracting metrics, and determining differences to adjust the autonomous vehicle's software for improved performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If autonomous vehicle control software is tested using real-world driving, then validation effectiveness is improved, but safety risks and costs increase
Solution Approach 1:
The patent creates virtual copies of real-world driving scenarios by recording actual vehicle trajectories, sensor data, and environmental conditions, then replaying these scenarios in a simulated environment. This allows comprehensive software validation without exposing real vehicles to safety risks, as the simulation reproduces critical driving situations including edge cases and hazardous conditions that would be dangerous to test in the real world.
Solution Approach 2:
The system performs preliminary testing and validation in the virtual environment before deploying software to real vehicles. By pre-testing software updates, edge cases, and safety-critical scenarios in simulation, the patent identifies and resolves issues before they could affect real-world safety, thereby improving validation effectiveness while minimizing actual safety risks.
2Object-affected harmful factors
If autonomous vehicle software is tested in simulated environments, then safety risks are reduced, but validation effectiveness deteriorates
Solution Approach 1:
The patent enhances simulation fidelity by copying actual vehicle log data including precise trajectories, sensor readings, environmental conditions, and human driver behaviors. This creates highly realistic virtual scenarios that accurately reproduce real-world driving conditions, maintaining validation effectiveness while enabling safe simulated testing of safety-critical software functions.
Solution Approach 2:
The system incorporates feedback mechanisms where simulation results are compared against expected real-world performance, and the virtual environment is continuously refined based on actual vehicle data. This feedback loop ensures the simulation accurately reflects real-world physics, vehicle dynamics, and environmental interactions, thereby maintaining high validation effectiveness despite using simulated rather than real-world testing.
3Reliability
If comprehensive logging and simulation are performed, then software validation quality is improved, but computational resources and time increase
Solution Approach 1:
The patent segments the comprehensive validation process into distinct phases: data collection during actual driving, data processing and scenario reconstruction, simulation execution, and result analysis. This segmentation allows parallel processing of different scenario types, prioritization of critical test cases, and efficient resource allocation, thereby maintaining high validation quality while reducing overall testing time through structured workflow optimization.
Data Source
AI summary
A simulation may be used to determine a difference between progress of a manually-driven vehicle and progress of a simulated autonomous vehicle. The method includes retrieving log data collected for the manually-driven vehicle driving along a route, generating a plurality of path segments for a portion of the route. The plurality of path segments corresponds to points in a lane that the manually-driven vehicle traveled through on the portion of the route. The method also includes running, using a software of the autonomous vehicle, a simulation of the autonomous vehicle driving along the plurality of path segments, extracting metrics from the log data and the simulation, and determining the difference between a first progress of the manually-driven vehicle and a second progress of the simulated autonomous vehicle based on the metrics.