Radar Object Classification via Detection Zone Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems struggle to accurately classify objects in a motor vehicle's lateral surrounding area, particularly distinguishing between stationary infrastructure objects like crash barriers and moving vehicles, leading to unnecessary warnings.
Innovation Solution
A method using a vehicle-mounted sensor device that segments the monitoring area into multiple detection areas, allowing for precise classification of objects based on signal components received from different angles, determining object type and distance, and employing radar technology to differentiate between stationary and moving objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single sensor device is used to monitor the lateral surrounding area, then the device complexity is reduced, but the object classification accuracy deteriorates
Solution Approach 1:
The monitoring area is segmented into multiple detection areas (first detection area for oblique angles, second detection area for perpendicular angles). This allows a single sensor device to function as multiple virtual sensors by processing signals from different spatial zones differently, thereby maintaining classification accuracy while using minimal hardware.
Solution Approach 2:
Different detection areas are assigned different evaluation criteria: the first detection area uses radial speed matching for stationary object detection, while the second detection area uses distance comparison for elongated object detection. This local differentiation of processing quality enables accurate classification using a single sensor device.
2Measurement precision
If the monitoring area is segmented into multiple detection areas with different evaluation criteria, then the object classification accuracy is improved, but the signal processing complexity increases
Solution Approach 1:
The signal processing is segmented by detection area, with each area having its own evaluation criteria. The first detection area processes signals using radial speed comparison, while the second detection area uses distance comparison. This segmentation enables accurate classification while keeping each processing module relatively simple and modular.
3Measurement precision
If radial speed matching is used to identify stationary objects, then the classification accuracy for stationary objects is improved, but false warnings may still occur for moving vehicles
Solution Approach 1:
The detection space is segmented into two zones with different evaluation methods. The first detection area (oblique angles) uses radial speed matching to identify stationary objects, while the second detection area (perpendicular angles) uses distance comparison to detect elongated objects. This spatial segmentation ensures that moving vehicles, which would show significant distance variations in the second area, are not misclassified as stationary objects.
Solution Approach 2:
The second detection area acts as an intermediary verification layer. Even if an object is initially identified as stationary based on radial speed in the first detection area, it must also satisfy the distance comparison criterion in the second detection area to be confirmed as an elongated stationary object, thereby filtering out false positives from moving vehicles.
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
Enables reliable classification of elongated stationary objects, such as crash barriers, while minimizing false warnings by accurately distinguishing them from moving vehicles, using a single sensor device and signal, thereby improving the accuracy and precision of object classification in the lateral surrounding area.
Implementation Method 1
a sensor device on the vehicle for monitoring the lateral surrounding area transmits a signal into the lateral surrounding area and the signal reflected on the object is received
Implementation Method 2
the signal reflected on the object is received
Implementation Method 3
A radial speed of the object is detected on the basis of the signal component of the signal received from the first detection area
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method for classifying an object (6) in a lateral surrounding region (5) of a motor vehicle (1). A signal is transmitted into the lateral surrounding region (5) by a vehicle-side sensor device (3) in order to monitor the lateral surrounding region (5), and the signal which is reflected on the object (6) is received. A monitoring region (E) of the sensor device (3) for monitoring the lateral surrounding region (5) is segmented into a first detection region (E1) and a second detection region (E2). A first value of a distance (AL) from the object (6) to the motor vehicle (1) is determined using the signal received from the first detection region (E1), and it is ascertained whether the object (6) is a stationary object located laterally to the motor vehicle (1). A second value of the distance (AL) from the object (6) to the motor vehicle (1) is ascertained using the signal received from the second detection region (E2), and the first value and the second value of the distance (AL) are used to determine whether the object (6) determined to be a stationary object is a stationary object which is elongated in the vehicle longitudinal direction (RL). The invention further relates to a driver assistance system (2) and to a motor vehicle (1).