Lane Keeping Trigger Logic Using Predicted Evasive Trajectory
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advanced driver assistance systems (ADAS) face challenges in accurately triggering lane keeping maneuvers due to the use of large lookup tables and unitless calibration parameters, leading to inconsistent triggering logic and reduced driving safety.
Innovation Solution
A computer-implemented method that determines the traveling trajectory of a host vehicle and detects lane or road boundaries to assess triggering conditions, allowing for precise calibration and performance of lane keeping maneuvers without relying on lookup tables, using measurable parameters with units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lookup tables and scaling factors are used to decide lane keeping aid triggering, then sufficient events can be covered to improve driving safety, but large lookup tables requiring additional memory are needed
Solution Approach 1:
The patent extracts and removes the large lookup tables from the system, replacing them with a geometric triggering logic that calculates lane boundary crossing events directly from sensor data and vehicle trajectory. This eliminates the need for storing extensive pre-computed event tables while maintaining comprehensive event coverage for safe lane keeping assistance.
Solution Approach 2:
The patent replaces the data-intensive lookup table approach with a computational geometric model that uses mathematical calculations based on vehicle position, velocity, and trajectory. This substitution transforms the system from memory-intensive table lookups to computation-intensive geometric calculations, reducing memory requirements while maintaining triggering accuracy.
2Ease of operation
If lookup tables and scaling factors with many calibration parameters are used, then triggering decisions can be made, but the calibration process becomes hard and validation becomes more difficult
Solution Approach 1:
The patent changes the parameter representation from unitless scaled values requiring extensive calibration to physical quantities with explicit units (meters, angles, velocities). The triggering conditions are expressed as geometric relationships between measurable physical parameters, making the system inherently more interpretable and easier to validate without requiring complex calibration procedures.
Solution Approach 2:
The patent enables the system to self-calibrate by using directly measurable physical quantities from sensors. The geometric triggering logic automatically adapts to different vehicle configurations and sensor characteristics through direct physical measurements, eliminating the need for manual calibration of numerous scaling parameters and making validation straightforward through physical verification.
3Ease of operation
If unitless parameters are used for calibration, then triggering decisions can be made, but the parameters make calibration harder and reduce driving safety
Solution Approach 1:
The patent fundamentally changes from using unitless scaled parameters to using physical parameters with explicit units (meters for distance, radians for angles, m/s for velocity). The triggering conditions are defined in terms of these physical quantities, making the system more transparent, easier to validate, and safer because the physical meaning of each parameter is explicit and can be directly verified against real-world measurements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Prior are approaches use peripheral data and lookup tables to decide if a lane keeping aid should be triggered. Such approaches require large lookup tables to cover all conceivable scenarios. Such approaches are difficult to calibrate, tune or validate. The present invention provides a method for determining a lane/road keeping maneuver of a host vehicle by use of at least one determined traveling trajectory of the host vehicle, a detected lane/road boundary and at least one triggering condition for performing the lane/road keeping maneuver. As a result, parameters for determining the lane/road keeping maneuver use units and are fewer in number in comparison to parameters of a lookup table, making the determining of the lane/road keeping maneuver more intuitive, more robust, easier to tune/calibrate and more secure since the system can be validated more easily.