Ramp Target Selection Using Lane-Aware Avoidance Driving Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current advanced driving systems for autonomous and semi-autonomous vehicles struggle with target selection on ramps, particularly those with one lane, as they rely on predictions and curvature calculations, leading to uncertainty and improper categorization of vehicles in shoulder areas.
Innovation Solution
The system determines the number of lanes on a ramp from a stored map and adjusts target selection accordingly, prioritizing the first vehicle ahead without relying on predictions or curvature calculations, using a single-lane scheme if necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If standard driving schemes prioritize targets directly in front of the host vehicle, then target selection is simple, but it fails to properly handle single-lane ramp situations where the target may not be directly in front
Solution Approach 1:
The system applies different target selection schemes based on the local characteristics of the roadway. When a ramp with a single lane is detected, a specialized target selection scheme is applied that prioritizes the first vehicle ahead regardless of its lateral position, rather than the standard scheme that prioritizes vehicles directly in front. This local adaptation resolves the contradiction by maintaining simplicity while improving accuracy for the specific ramp scenario.
2Measurement precision
If predictions and curvature calculations are used to select targets on ramps, then target selection may be more accurate, but uncertainty and calculation complexity increase
Solution Approach 1:
The patent extracts and removes the complex prediction and curvature calculation components from the target selection process on single-lane ramps. Instead of using these complex methods, the system simply identifies the first vehicle ahead in the single lane. This extraction eliminates the source of uncertainty and calculation complexity while maintaining sufficient accuracy for the specific single-lane ramp scenario.
3Adaptability or versatility
If the system uses different target selection schemes for different lane configurations, then adaptability to various ramp scenarios is improved, but system complexity increases
Solution Approach 1:
The system dynamically selects the appropriate target selection scheme based on the detected lane configuration. When a single-lane ramp is detected, the simplified scheme is activated; when multiple lanes are detected, the standard scheme is used. This dynamic adaptation allows the system to maintain high versatility across different scenarios while managing complexity by only activating the more complex scheme when necessary.
Data Source
AI summary
This document describes techniques and systems for target of interest selection on roadway ramps. When a host vehicle transitions from a roadway to a ramp, a quantity of lanes of the ramp is determined from a map stored in a driving system of the host vehicle. If there are multiple lanes, then a standard driving scheme may be employed. This standard driving scheme may prioritize a target directly in front of the host vehicle to be a target of interest. However, if the ramp has a single lane, a first target leading the host vehicle, but not necessarily directly in front of the host vehicle is determined to be the target of interest. Driving systems, such as adaptive cruise control, may base driving decisions on the target of interest. By determining the target of interest in this manner, uncertainty due to predictions and complexity of calculations may be reduced.