Vehicle Roll and Pitch Estimation From Urban Structure Edges
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining the roll angle and pitch angle of a means of transport, such as motor vehicles, face challenges due to sensor drift and limited accuracy in urban environments, where traditional sensors like GNSS antennas and level sensors are unreliable or costly, and visual horizon recognition is not applicable for land vehicles.
Innovation Solution
A method that detects environmental information using sensors like cameras, lidar, or laser scanners to determine the edges of artificial structures, utilizing a database for edge orientation data to calculate the roll and pitch angles, allowing for precise angle measurement and improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional sensors (acceleration sensors, yaw rate sensors) are used to determine roll angle and pitch angle, then the measurement can be performed using available sensors, but the accuracy deteriorates due to sensor drift and indirect measurement limitations
Solution Approach 1:
The patent introduces environmental information (images of artificial structures) as an intermediary to determine vehicle orientation. Instead of directly measuring angles with drift-prone sensors, the system captures images of known structures, detects their edges, and uses the expected geometric orientations of these edges to calculate roll and pitch angles, thereby eliminating sensor drift issues
Solution Approach 2:
The patent replaces the mechanical sensor-based measurement system (acceleration sensors and yaw rate sensors) with an optical/image processing system. By substituting physical sensors that suffer from drift with camera-based environmental observation and computational geometry, the system achieves more reliable and precise angle measurements
2Measurement precision
If GNSS antennas are used for attitude measurement, then the measurement can be performed, but the cost increases considerably
Solution Approach 1:
The patent makes existing environmental sensors (cameras, lidar) perform multiple functions: they continue to capture environmental data for navigation and positioning purposes while simultaneously enabling roll and pitch angle determination through edge detection of artificial structures. This eliminates the need for separate expensive attitude measurement devices like GNSS antennas
Solution Approach 2:
The patent uses a copy or representation of the physical environment (digital images of artificial structures) to determine vehicle orientation. By processing visual information about known structures rather than using direct physical measurement devices, the system achieves attitude measurement without the cost of specialized hardware
3Ease of operation
If level sensors in the chassis are used to measure wheel suspension level, then the roll and pitch angles can be estimated, but the accuracy is insufficient
Solution Approach 1:
The patent transitions from one-dimensional local measurements (wheel suspension level at discrete points) to two-dimensional or three-dimensional environmental observation (edges of artificial structures in the visual field). By observing the orientation of extended structures in space rather than point measurements, the system achieves superior angle estimation accuracy
4Measurement precision
If visual horizon recognition is used for roll and pitch estimation, then the angles can be determined, but the method is not applicable for land vehicles
Solution Approach 1:
The patent adapts visual orientation methods by changing the reference from the global horizon (inapplicable for land vehicles) to local artificial structures with known geometric properties. By focusing on the specific local features of the environment (edges of buildings, poles, etc.) rather than the distant horizon, the method becomes applicable to land vehicles while maintaining precision
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The present invention relates to a method, a computer program with instructions, and a device for determining the roll angle or pitch angle of a means of transportation, in particular a motor vehicle. In a first step, environmental information for the environment of the means of transportation is acquired (S1). Edges of artificial structures are then determined from the environmental information (S2). Optionally, the orientation of the edges of the artificial structures can be determined from a database of edge data (S3). Finally, the roll angle or pitch angle of the means of transportation is determined based on the edges (S4).