Virtual Stop Line Control for Crosswalk Pedestrian Deceleration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in safely navigating crosswalks, as existing systems often result in rapid deceleration or insufficient deceleration due to inaccurate pedestrian recognition and cognitive information, leading to safety and ride comfort issues.
Innovation Solution
A method and apparatus for generating a virtual stop line by determining a target crosswalk, setting a region of interest, and deciding whether to generate a virtual stop line based on pedestrian location, using sensors and processors to recognize pedestrians and estimate their intentions, thereby optimizing vehicle speed and trajectory planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the vehicle uses existing pedestrian recognition systems to navigate crosswalks, then the vehicle can detect pedestrians, but the system causes rapid deceleration or insufficient deceleration due to inaccurate pedestrian recognition and cognitive information
Solution Approach 1:
The system generates a virtual stop line in advance before the actual crosswalk location based on predicted pedestrian behavior and vehicle trajectory. This preliminary action allows the vehicle to prepare for deceleration earlier and more smoothly, avoiding sudden braking while ensuring safety. The virtual stop line serves as a proactive control reference point that anticipates the need to stop.
Solution Approach 2:
The virtual stop line acts as an intermediary element between the pedestrian detection system and the vehicle's braking system. Instead of directly translating pedestrian detection into deceleration commands, the system introduces this virtual reference line that mediates the control process, enabling smoother and more reliable deceleration based on the spatial relationship between the vehicle and the virtual stop line.
2Ease of operation
If the vehicle generates a virtual stop line based on accurate pedestrian location, then ride comfort is improved by preventing rapid deceleration, but the system complexity increases due to additional processing requirements
Solution Approach 1:
The existing pedestrian detection and cognitive information processing systems are extended to serve dual purposes: their original function of detecting pedestrians is maintained, and they now also provide data for generating the virtual stop line. This multi-functionality approach allows the system to improve ride comfort through virtual stop line generation without requiring entirely separate hardware or processing systems.
Solution Approach 2:
The virtual stop line is essentially a computational copy or representation of where the vehicle should stop, created in the digital processing domain rather than physical space. This copying approach allows the system to work with simplified geometric representations and predictions rather than complex real-time physical measurements, reducing the computational burden while improving comfort.
3Reliability
If the vehicle uses traditional stop line generation methods, then the system is simpler to implement, but safety is compromised due to insufficient deceleration control
Solution Approach 1:
By generating the virtual stop line in advance based on predicted pedestrian behavior and vehicle trajectory, the system ensures safety is maintained while enabling smoother deceleration. The preliminary generation of this control reference allows the vehicle to approach the crosswalk at a controlled rate, preventing both rapid and insufficient deceleration scenarios that compromise safety.
Data Source
AI summary
Provided are a method and apparatus for generating a virtual stop line, wherein the method includes determining a target crosswalk from among one or more crosswalks on a driving route, setting a region of interest for the target crosswalk, and determining whether or not to generate a virtual stop line, on the basis of a location of a pedestrian within the region of interest.