ADAS Feature Availability Layers for Sensor Uncertainty Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face malfunctions and safety risks due to unreliable sensor data, leading to potential collisions or property damage when required sensor data is unavailable or insufficient for advanced autonomous functions.
Innovation Solution
A method to determine uncertainty levels for driving data streams, generating a total uncertainty level, and disabling or enabling autonomous functions based on predefined availability criteria and thresholds to prevent unsafe operations, using techniques like weighted averaging, multiplication, or summation, and projecting data into an occupancy grid to create an uncertainty map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If autonomous functions are enabled with high data requirements, then functionality and advanced features are improved, but safety and reliability deteriorate when sensor data is unavailable or insufficient
Solution Approach 1:
The patent segments autonomous functions into multiple layers based on data requirements and uncertainty levels. Basic functions operate at lower layers with minimal data requirements, while advanced functions operate at higher layers requiring robust multi-sensor data. This segmentation allows the system to provide appropriate functionality while maintaining safety by preventing advanced functions from operating when their data requirements are not met.
Solution Approach 2:
The system dynamically adjusts which autonomous functions are available based on real-time sensor data quality and uncertainty levels. The availability of specific autonomous functions is not fixed but changes according to the current operational context and data reliability, allowing the system to optimize between functionality and safety conditions.
2Reliability
If uncertainty thresholds are set low to ensure safety, then collision risk is reduced, but functional availability and productivity deteriorate
Solution Approach 1:
The patent applies segmentation by creating multiple uncertainty thresholds corresponding to different function layers. Instead of a single low threshold that would disable all functions, the system has basic functions with lower thresholds and advanced functions with higher thresholds. This allows basic functions to remain available while advanced functions are selectively enabled only when their specific data requirements are satisfied.
Solution Approach 2:
The system changes the uncertainty threshold parameter dynamically based on the specific autonomous function being evaluated. Different functions have different threshold parameters tailored to their data requirements, allowing the system to maintain safety while maximizing functional availability by not applying a uniformly restrictive threshold.
3Measurement precision
If multiple sensor data streams are integrated for robust autonomous functions, then navigation accuracy is improved, but system complexity and computational requirements increase
Solution Approach 1:
The patent segments the data integration process into layers, where basic navigation functions use only essential sensor data streams, and advanced functions progressively integrate additional streams. This segmentation allows the system to achieve high navigation accuracy when needed while avoiding the complexity of processing all possible data streams for all functions.
Solution Approach 2:
The system applies partial action by integrating only the necessary subset of sensor data streams for each specific autonomous function rather than all available streams. Basic functions use partial data integration, while advanced functions use more comprehensive integration only when required, avoiding excessive computational complexity.
Data Source
AI summary
A method of controlling availability of autonomy functions of a vehicle includes determining one or more uncertainty levels, each corresponding to a driving data stream including driving data, wherein the uncertainty levels correspond to a probability of a collision or driving off the roadway. The method further includes generating a total uncertainty level based on the one or more uncertainty levels and disabling a first autonomy function based on availability criteria of the first autonomy function. The availability criteria define a first threshold of the total uncertainty level. The method further includes providing a second autonomy function based on the second autonomy function's availability criteria that include a total uncertainty threshold higher than the first threshold.