Intersection Stop Location Detection with Uncertainty Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving vehicles face challenges in determining appropriate stopping locations at intersections, especially when there are no painted stop lines, due to changes in road conditions or markings, which can affect the ability to detect approaching vehicles and other objects.
Innovation Solution
A method using a neural network to analyze sensor data, including lidar point clouds and camera images, to generate heatmaps and vectors for stop line detection, and a localization channel to determine the most likely stop location, incorporating uncertainty estimation and roadgraph features to predict and update stop locations in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the vehicle relies on pre-mapped stop line locations, then navigation efficiency is improved, but reliability deteriorates when stop lines change due to construction or road wear
Solution Approach 1:
The system performs preliminary mapping of stop lines during normal operation, storing this information for efficient navigation. However, it also continuously monitors for changes in stop line locations during real-time operation, allowing it to detect and adapt to construction zones or road wear that may have altered the stop line position since mapping.
Solution Approach 2:
The system uses feedback from multiple sensors (cameras, LIDAR) to continuously verify the actual stop line location against the pre-mapped location. When discrepancies are detected, the system updates its understanding of the stop line position, ensuring reliable navigation even when road conditions change.
2Reliability
If the vehicle stops earlier to ensure safety, then detection of approaching vehicles is improved, but loss of time increases
Solution Approach 1:
The system dynamically adjusts the stopping position based on real-time detection of approaching vehicles and pedestrians. When objects are detected in the intersection, the system automatically extends the stopping position forward to improve detection capability. When the path is clear, it maintains the optimal stop line position to minimize stopping time.
3Measurement precision
If multiple sensors and processing channels are used to improve stop location detection, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system segments the stop line detection task into three specialized processing channels: a classification channel that identifies whether a stop line is present, a localization channel that determines the precise location, and an uncertainty estimation channel that quantifies confidence. This segmentation allows each channel to be optimized for its specific function while working together to achieve high overall precision.
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.