Vehicle Function Verification Under Perception Failure Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The performance of perception algorithms in autonomous and assisted driving systems is affected by hardware and environmental changes, leading to failures in meeting expected functional safety and performance requirements.
Innovation Solution
A method and apparatus for verifying vehicle functions by simulating various perception failure situations, allowing for the determination of the tolerance of vehicle functions to perception algorithm performance and guiding the development and optimization of the perception algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If perception algorithms are used for autonomous driving functions, then the vehicle can provide assisted driving services (AEB, ACC, NOA, HWP), but the performance is affected by hardware and environmental changes causing failures to meet functional safety requirements
Solution Approach 1:
The patent performs perception failure simulation in advance before actual deployment. By pre-simulating various perception failures (missing detection, false detection, detection errors) under different environmental conditions and hardware configurations, the system establishes baseline safety margins and identifies critical failure modes beforehand, enabling proactive mitigation strategies
Solution Approach 2:
The patent applies preliminary anti-action by introducing artificial perception failures into the simulation to counteract potential real-world failures. By injecting detection errors, false positives, and sensor failures in the virtual environment, the system prepares the vehicle control system to handle these adverse conditions when they occur in reality
Solution Approach 3:
The patent implements feedback mechanisms by continuously monitoring verification results from perception failure simulations. The system uses this feedback to iteratively optimize the perception algorithm, adjust safety margins, and refine the assisted driving control strategies to maintain functional safety compliance across varying conditions
2Measurement precision
If perception algorithms rely on sensor data (vision, laser radar, ultrasonic radar, millimeter wave radar), then the vehicle can perceive environmental objects, but hardware changes and environmental factors cause perception performance degradation
Solution Approach 1:
The patent applies parameter changes by systematically varying sensor parameters (detection ranges, detection angles, detection frequencies) and environmental parameters (weather conditions, lighting conditions, road conditions) in simulations. This allows the system to identify optimal parameter configurations that maintain measurement precision across different hardware and environmental conditions
Solution Approach 2:
The patent implements universality by developing a comprehensive perception failure simulation framework that works across multiple sensor types (vision, laser radar, ultrasonic radar, millimeter wave radar) and various assisted driving functions (AEB, ACC, NOA, HWP). The simulation system universally handles different failure modes and environmental conditions to verify functional safety across the entire assisted driving system
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Embodiments of the present disclosure disclose a method and apparatus for verifying a vehicle function, an electronic device, and a storage medium, wherein the method includes: acquiring operation scenario data respectively corresponding to each of one or operation scenarios; determining, based on the operation scenario data respectively corresponding to each of the operation scenarios, first perception result information corresponding to a preset vehicle function of an ego vehicle in each of the operation scenarios; performing perception failure simulation processing on the first perception result information based on a perception failure simulation rule to obtain second perception result information corresponding to each of the operation scenarios; and verifying the preset vehicle function based on the second perception result information to obtain a verification result.