Autonomous Vehicle Camera Signal Qualification Using Map-Based Checks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing camera systems in partially or fully autonomous vehicles lack safety requirements, leading to potential accidents due to unrecognized hardware or software faults in image/camera signals, which are not adequately qualified to meet ASIL A-D or SIL 1-3 standards.
Innovation Solution
A method that monitors camera signals for recurring time windows, sequence, corruption, and signature checking, and compares them with infrastructure information using a map check function to qualify image/camera signals, ensuring they meet safety standards by identifying and marking faulty signals, thereby preventing continued calculation with incorrect data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If camera systems without safety requirements are used in autonomous vehicles, then device complexity and cost are reduced, but reliability and safety are compromised due to unrecognized hardware or software faults
Solution Approach 1:
The patent implements preliminary qualification of camera systems by comparing their output signals against pre-stored infrastructure information (maps, traffic lights, road signs) before the signals are used for autonomous driving decisions. This advance verification ensures that only reliable signals from properly functioning cameras are processed, preventing faults from compromising safety while allowing the use of standard camera systems
2Reliability
If comprehensive monitoring and qualification of camera signals is implemented, then reliability and safety are improved, but device complexity and processing requirements increase
Solution Approach 1:
The patent introduces an intermediary qualification unit that acts as a mediator between the camera system and the autonomous driving control system. This unit performs signal qualification by comparing camera outputs against infrastructure information databases, filtering out faulty signals before they reach the decision-making systems. This intermediary approach maintains reliability while managing complexity through modular architecture
3Measurement precision
If real-time verification of camera signals against infrastructure information is performed, then measurement precision and detection accuracy are improved, but processing time and computational load increase
Solution Approach 1:
The patent performs preliminary comparison of camera signals against pre-stored infrastructure information (maps, known locations of traffic lights and road signs) before final object detection. By having infrastructure data readily available and performing preliminary verification, the system achieves high measurement precision without excessive processing delays during critical driving moments
Data Source
AI summary
A method for qualification of a camera system and/or its image/camera signals of a partially autonomously or autonomously driving vehicle. Infrastructure information from a vehicle's surroundings is used. The image/camera signal of the camera system is monitored by a monitoring function with regard to at least the following criteria: image/camera signal in recurring, known time windows, sequence of the image/camera signals, corruption of the image/camera signals, signature checking of the camera system. Checking of the image/camera signals using unmoving land markings that are reconciled with a list is carried out and they are checked for a mapping signature. Then the image/camera signals and the infrastructure information contained in the updated map check function are communicated to an evaluation function and merged and compared in a signal fusion unit. Consequently, a qualified image/camera signal is obtained.