Static Radar Target Classification Using Occupancy Grids
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar sensor systems for motor vehicles struggle to reliably distinguish between static obstacles and drivable objects, often treating them as point-like targets, which can lead to inaccurate classification and increased false-positive obstacle detections.
Innovation Solution
The method involves assigning an occupancy pattern to static radar targets based on their radar signatures, which is then entered into an occupancy grid. This pattern indicates occupancy probabilities for specific grid positions, allowing for the estimation of the extent of static obstacles and enabling more accurate classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If radar targets are treated as point-like objects, then the device complexity is reduced, but the measurement precision of object extent is deteriorated
Solution Approach 1:
The patent segments the occupancy space into a grid structure, where each cell can be independently occupied or unoccupied. This segmentation allows the system to represent objects not as single points but as collections of occupied grid cells, thereby improving extent estimation while maintaining manageable complexity through discrete representation.
Solution Approach 2:
The patent transitions from representing objects in one dimension (point targets with coordinates) to two dimensions (occupancy grids with row and column indices). This dimensional expansion enables the representation of object extent and shape by mapping radar reflections onto a grid, allowing distinction between point targets and extended objects.
2Reliability
If occupancy patterns are assigned to static radar targets, then the reliability of obstacle detection is improved, but the device complexity is worsened
Solution Approach 1:
The occupancy grid serves multiple functions simultaneously: it represents the environment layout, tracks object positions, estimates object extent through occupancy patterns, and enables classification of static targets. This multi-functionality improves reliability without proportionally increasing complexity, as a single data structure accomplishes multiple tasks.
Solution Approach 2:
The patent creates an abstract copy of the physical environment in the form of an occupancy grid. This virtual representation copies the essential spatial information from radar measurements, allowing the system to analyze and classify objects without directly manipulating complex raw radar data, thus improving reliability while managing complexity through abstraction.
3Measurement precision
If radar reflections are accumulated over time, then the measurement precision of object classification is improved, but the loss of time is worsened
Solution Approach 1:
The system performs preliminary actions by continuously accumulating radar reflections and updating the occupancy grid in advance. This preparation allows the classification process to work with pre-processed occupancy patterns rather than raw radar data, improving classification precision while reducing the computational time required during actual evaluation through pre-computed occupancy information.
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 reliability of static obstacle detection by allowing for the accurate estimation of object extent and classification, reducing false-positive detections and improving the mapping of the vehicle's surroundings.
Implementation Method 1
a radar sensor for motor vehicles having: a radar signal transmitting unit configured to emit radar signals in a detection area in front of the motor vehicle; a radar signal receiving unit configured to receive radar signals reflected by objects in the detection area
Implementation Method 2
receive radar signals reflected by objects in the detection area
Data Source
Figure 1~2(d)
Figure 3
AI summary
The invention relates to a method for identifying and classifying static radar targets by means of a radar sensor of a motor vehicle, comprising: identifying (S10) an object (14) as a static radar target on the basis of the received radar signals reflected by the object (14), producing (S12) an occupancy pattern (24) in an occupancy grid (20) on the basis of the received radar signals reflected by the object (14), storing (S14) an association that associates the produced occupancy pattern (24) with the static radar target, classifying (S16) the static radar target as belonging to one of a plurality of groups of static radar targets on the basis of characteristic features of radar signatures of the received radar signals reflected by the corresponding object (14). The invention further relates to a radar sensor for carrying out the method.