Intersection Stop Location Detection for Changing Stop Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving vehicles face challenges in determining appropriate stopping locations at intersections, especially when painted stop lines are absent or have changed due to construction or wear and tear, which can impact the detection of approaching vehicles and objects.
Innovation Solution
A method using onboard processors to evaluate training data and generate classification, localization, and uncertainty estimation channels to identify and predict stop line locations, incorporating sensor data from lidar and camera images, and projecting stop line points onto a lane tangent to determine a predicted stopping location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the vehicle relies on pre-mapped stop line locations, then the stopping process is simple and efficient, but the stop line location may change due to construction or road resurfacing, leading to inaccurate stopping
Solution Approach 1:
The system performs preliminary detection of stop line locations using sensor data (cameras, lidars) before reaching the intersection. It evaluates multiple candidate stop line locations and prepares stop candidate locations in advance, allowing the vehicle to adapt if the pre-mapped stop line has changed or is no longer valid.
Solution Approach 2:
The system continuously monitors the actual stop line location using sensor data and compares it with the pre-mapped stop line location. Based on this feedback, it dynamically adjusts the stop candidate locations to ensure accurate stopping even when road conditions change due to construction or resurfacing.
2Measurement precision
If the vehicle stops exactly at the stop line, then stopping precision is maximized, but the ability to detect approaching vehicles and objects along cross streets is reduced
Solution Approach 1:
The system generates multiple stop candidate locations rather than a single stop point. Each candidate location has different qualities - some are closer to the stop line for precise stopping, while others are positioned to optimize detection of approaching vehicles. The system selects the most appropriate candidate based on real-time conditions.
Solution Approach 2:
The system dynamically adjusts the selected stop candidate location based on detected objects and approaching vehicles. If approaching vehicles or pedestrians are detected, the system may select a stop candidate location that provides better visibility and safety, rather than strictly adhering to the stop line position.
3Difficulty of detecting and measuring
If the vehicle stops earlier to improve detection capability, then detection of approaching vehicles is enhanced, but stopping precision at the correct location deteriorates
Solution Approach 1:
The system segments the stopping decision into multiple stop candidate locations rather than a single fixed point. This segmentation allows the vehicle to evaluate different stopping positions and select the one that best balances detection capability and stopping precision based on real-time environmental conditions.
Data Source
AI summary
The technology relates to approaches for determining appropriate stopping locations at intersections for vehicles operating in a self-driving mode. While many intersections have stop lines painted on the roadway, many others have no such lines. Even if a stop line is present, the physical location may not match what is in store map data, which may be out of date due to construction or line repainting. Aspects of the technology employ a neural network that utilizes input training data and detected sensor data to perform classification, localization and uncertain estimation processes. Based on these processes, the system is able to evaluate distribution information for possible stop locations. The vehicle uses such information to determine an optimal stop point, which may or may not correspond to a stop line in the map data. This information is also used to update the existing map data, which can be shared with other vehicles.