Vehicle Curb Detection Using Spatial Lines and Separation Points
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in accurately determining curb locations for navigation and user drop-off/pick-up proximity without relying on potentially inaccurate maps, especially in dynamic environments like construction zones.
Innovation Solution
The vehicle generates sensor data using lidar, radar, or other sensors to identify points associated with driving surfaces and sidewalks, then analyzes this data to generate spatial lines and separation points, which are used to create a curve representing the curb, allowing for real-time curb detection and navigation adjustments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the autonomous vehicle relies on maps to determine curb locations, then the navigation process is simplified, but the accuracy and reliability of curb detection deteriorates in dynamic environments like construction zones
Solution Approach 1:
The system segments the curb detection task into multiple components: sensor data acquisition, spatial line generation, separation point identification, and curve fitting. This segmentation allows real-time processing of complex environmental data while maintaining high detection accuracy without relying on pre-stored maps.
Solution Approach 2:
The patent replaces the mechanical reliance on static maps with a sensor-based perception system using lidar and cameras. This substitution enables dynamic curb detection by directly sensing the physical environment, thereby improving reliability in construction zones and other dynamic environments where maps may be outdated or inaccurate.
2Reliability
If the vehicle uses sensor data processing to detect curbs in real-time, then the reliability of curb detection improves, but the computational complexity and processing time increase
Solution Approach 1:
The system performs preliminary actions by pre-processing sensor data to generate spatial lines and identify separation points before final curb detection. This preliminary processing organizes raw sensor data into structured spatial representations, reducing the computational burden during real-time curb detection and enabling faster processing despite the complexity of the task.
Solution Approach 2:
The patent implements dynamic processing where the system adapts its detection parameters and processing intensity based on environmental conditions. In complex environments like construction zones, the system dynamically adjusts processing to maintain high reliability while managing computational resources efficiently, balancing accuracy with processing speed.
3Measurement precision
If the vehicle processes detailed sensor data to generate spatial lines and separation points, then the precision of curb location determination improves, but the processing time and computational resources increase
Solution Approach 1:
The system performs preliminary generation of spatial lines from sensor data points before conducting separation point analysis. This preliminary structuring of data into spatial lines reduces the dimensionality of the problem and enables faster identification of curb locations while maintaining high measurement precision through the systematic analysis of spatial relationships.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method enables accurate and dynamic curb detection, improving navigation safety and proximity to drop-off/pick-up locations without relying on maps, enhancing the vehicle's ability to adapt to changing environments.
Implementation Method 1
receiving sensor data from a lidar sensor, the sensor data representing first points associated with a drivable surface and second points associated with a sidewalk
Data Source
AI summary
Techniques for identifying curbs are discussed herein. For instance, a vehicle may generate sensor data using one or more sensors, where the sensor data represents points associated with a driving surface and a sidewalk. The vehicle may then quantize the points into distance bins that are located laterally along the driving direction of the vehicle in order to generate spatial lines. Next, the vehicle may determine separation points for the spatial lines, where the separation points are configured to separate the points associated with the driving surface from the points associated with the sidewalk. The vehicle may then generate, using the separation points, a curve that represents the curb between the driving surface and the sidewalk. This way, the vehicle may use the curve while navigating, such as to avoid the curb and/or stop at a location that is proximate to the curb.