Vehicle Sensor Fusion Monitoring for Object And Free-Space Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is a need for methods and systems that ensure the reliability of the perception system in Automated Driving Systems (ADS), as the quality of the sensor fusion output significantly affects the vehicle's perception capability and overall performance and safety.
Innovation Solution
A computer-implemented method and system for monitoring the reliability of a sensor fusion system in a vehicle. The method involves storing input data from vehicle-mounted sensors, generating perception output data, comparing the output data with the stored input data to validate object detections and free-space area indications, and outputting a signal indicative of the sensor fusion system's status based on validated objects and areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor fusion system processes multiple sensor inputs to generate perception output data, then the completeness and accuracy of environment understanding is improved, but the complexity of the system increases
Solution Approach 1:
The validation process is segmented into distinct modules: object detection validation, free-space area validation, and status determination. Each module independently validates specific aspects of the perception output data against corresponding sensor inputs, breaking down the complex validation task into manageable segments that can be processed separately and systematically.
Solution Approach 2:
The system introduces an intermediary validation mechanism that acts as a bridge between the raw sensor inputs and the perception output data. This intermediary layer compares and verifies the consistency between sensor data and perception results, ensuring accuracy without requiring direct modification of the core sensor fusion algorithms.
2Reliability
If the system validates perception output data against sensor input data, then the reliability of the sensor fusion system is improved, but the processing time and computational load increase
Solution Approach 1:
The system performs preliminary validation by comparing perception output data with sensor input data before the perception results are used for critical driving decisions. This advance validation ensures that unreliable perception data is identified early, preventing potential safety issues while maintaining efficient real-time operation.
Solution Approach 2:
The validation process focuses on critical aspects of the perception output data, such as object detections and free-space area indications, rather than validating every single data point. This partial validation approach ensures system reliability for safety-critical functions while avoiding excessive processing time that would result from comprehensive validation of all perception data.
3Object-affected harmful factors
If the system monitors and validates each sensor output, then the safety of ADS functions is improved, but the device complexity and computational resources required increase
Solution Approach 1:
The validation system applies different validation strategies and thresholds to different types of perception output data based on their safety criticality. Object detections receive validation focused on presence and position accuracy, while free-space area indications receive validation focused on boundary accuracy. This localized validation approach ensures appropriate safety monitoring without uniformly increasing complexity across all validation processes.
Data Source
AI summary
A method for monitoring a reliability of an output of a sensor fusion system of a vehicle is disclosed. At first, an input data, including sensor data obtained over a period of time from each of vehicle-mounted sensors, which are configured to monitor a surrounding environment of the vehicle, of the sensor fusion system is stored. Perception output data that is output from the sensor fusion system, includes object detections and free-space area indications in the surrounding environment of the vehicle. The obtained perception output data is then compared with the stored input data and determined whether any object detections indicated in the obtained perception output data is indicated in the field-of-view of that sensor in order to validate any object detections and free-space area indications in the obtained perception output data. Accordingly, a signal indicative of a status of the sensor fusion system is outputted.