Sensor Apparatus Spatial Density Object Categorization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle sensor systems face challenges in accurately distinguishing stationary objects, such as crash barriers, from non-stationary objects, like moving vehicles, due to incorrect tracking and categorization, especially when partial segments of stationary objects appear to move with the vehicle.
Innovation Solution
The method determines the spatial density of captured object points and compares it with a predetermined threshold value, using angular resolution and orientation of transmitted and received signals to categorize objects as stationary or non-stationary, thereby improving recognition accuracy by differentiating between stationary and non-stationary objects based on density and orientation relative to the sensor's main monitoring direction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If stationary objects are tracked using conventional sensor systems, then the system can detect objects in the monitoring region, but the objects are incorrectly categorized as non-stationary due to apparent motion caused by vehicle movement
Solution Approach 1:
The patent changes the parameter used for categorization from apparent motion detection to spatial density analysis. By calculating the density of object points in the captured spatial distribution and comparing it to a threshold, the system can correctly identify stationary objects regardless of their apparent motion, thus resolving the categorization error
Solution Approach 2:
The patent introduces a new dimension of analysis by considering the spatial distribution and density of object points rather than relying solely on temporal motion information. This dimensional shift from time-based to space-based analysis enables correct identification of stationary objects
2Productivity
If the sensor apparatus captures all objects in the monitoring region, then complete object detection is achieved, but differentiation between stationary and non-stationary objects becomes difficult
Solution Approach 1:
The patent introduces spatial density as a new parameter for object differentiation. By calculating the density of object points in the captured spatial distribution and comparing it to a predetermined threshold, the system can accurately distinguish stationary objects (higher density) from non-stationary objects (lower density) while maintaining complete detection
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 approach enhances the precision and efficiency of object categorization, preventing incorrect tracking of stationary objects as non-stationary and allowing for better differentiation between crash barriers and moving vehicles, thereby improving the functionality of driver assistance systems.
Implementation Method 1
transmitted signals reflected from object points of the at least one object are captured as received signals
Implementation Method 2
a distance and/or a relative speed of at least one object point of the at least one object can be determined in accordance with an in particular sampling light time-of-flight measurement method, in particular a LiDAR or a LaDAR method
Data Source
AI summary
A method is described for the in particular optical capture of at least one object (18, 20) with at least one sensor apparatus (14) of a vehicle (10), a device (34) of a sensor apparatus (14), a sensor apparatus (14) and a driver assistance system (12) with at least one sensor apparatus (14). In the method, in particular optical transmitted signals (36) are transmitted into a monitoring region (16) with the at least one sensor apparatus (14) and transmitted signals (36) reflected from object points (40) of the at least one object (18, 20) are captured as received signals (38) with angular resolution with reference to a main monitoring direction (42) of the at least one sensor apparatus (14). A spatial distribution of the object points (40) of the at least one object (18, 20) relative to the at least one sensor apparatus (14) is determined from a relationship between the transmitted signals (36) and the received signals (38), and the at least one object (18, 20) is categorized as stationary or non-stationary. A spatial density of the captured object points (40) in at least one region of the at least one object (20) is determined, and if the density of the captured object points (40) is smaller than a predetermined or predeterminable threshold value, the at least one object (20) is categorized as stationary.