Vehicular Sensor Data Correlation via Probability Density Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems face challenges in accurately correlating sensor data to generate a reliable model of objects in the vehicle's environment, particularly when dealing with uncertain information, which can lead to errors such as false positives and false negatives.
Innovation Solution
A system that combines data from multiple sensors using model generators and comparison controllers to create environmental models with probability density distributions, determining potential matchings and distance probability functions to enhance the accuracy and integrity of object detection, operating at a higher Automotive Safety Integrity Level (ASIL) through ASIL decomposition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor data from multiple sensors is combined using traditional sensor fusion algorithms, then the quality of environmental model is improved, but the system cannot achieve higher Automotive Safety Integrity Level due to handling of uncertain information
Solution Approach 1:
The patent changes the parameter representation from deterministic values to probability density distributions. Each object characteristic (position, velocity, etc.) is represented as a probability distribution rather than a single value, allowing the system to quantify and propagate uncertainty through the sensor fusion process while maintaining high ASIL standards.
Solution Approach 2:
The patent replaces traditional sensor fusion algorithms (which operate on deterministic values) with a probabilistic framework using probability density distributions. This substitution allows the system to handle uncertain information in a mathematically rigorous way that meets higher ASIL requirements while improving detection accuracy.
2Measurement precision
If probability density distributions are used to represent uncertain object characteristics, then the accuracy and reliability of object detection is improved, but the computational complexity and system integrity requirements increase
Solution Approach 1:
The patent segments the environmental model into multiple independent object representations, each with its own probability density distribution. This segmentation allows the system to manage complexity by treating each object separately while maintaining overall system accuracy through individual probabilistic modeling of each detected object's characteristics.
3Reliability
If multiple environmental models are generated and correlated, then the reliability and safety integrity level are improved, but the processing time and computational resources increase
Solution Approach 1:
The patent performs preliminary actions by generating multiple environmental models with probability density distributions before final correlation. This allows the system to pre-compute and organize uncertain information in a structured format, reducing the computational burden during real-time correlation and decision-making processes.
Data Source
AI summary
A system for correlating sensor data in a vehicle includes a first sensor disposed on the vehicle to detect a plurality of first objects. A first object identification controller analyzes the first data stream, identifies the first objects, and determines first characteristics associated therewith. A second sensor disposed on the vehicle detects a plurality of second objects. A second object identification controller analyzes the second data stream, identifies the second objects, and determines second characteristics associated therewith. A model generator includes a plausibility voter to generate an environmental model of the objects existing in space around the vehicle. The model generator may use ASIL decomposition to provide a higher ASIL level than that of any of the sensors or object identification controllers alone. Matchings between uncertain objects are accommodated using matching distance probability functions and a distance-probability voter. A method of operation is also provided.