Ultrasonic Echo Classification for Moving Pedestrian Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ultrasonic sensor systems struggle with unreliable detection of poorly reflective objects and objects in motion, particularly pedestrians, due to conflicting requirements in resolution and dynamic object identification, leading to incorrect classifications and increased computing time.
Innovation Solution
The method involves emitting ultrasonic signals at multiple time points, receiving echo signals, and classifying objects based on echo properties and distance changes, using direct and cross echoes to distinguish between static and dynamic objects, particularly pedestrians, by evaluating multiple reflections and relative speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If echo tracing method is used to detect moving objects, then the ability to detect moving reflection points on moving objects is improved, but the computing time increases significantly and robustness decreases due to incorrect echo assignment
Solution Approach 1:
The patent extracts only the necessary information (distance changes, velocity, acceleration) directly from echo signals without performing full echo tracing. By taking out only the essential parameters needed for classification rather than complete trajectory analysis, the system achieves reliable moving object detection with significantly reduced computing time.
Solution Approach 2:
The system performs preliminary classification of objects as static or dynamic based on echo signal properties before detailed analysis. This preliminary action filters out objects that don't require extensive processing, reducing overall computing time while maintaining detection reliability for moving objects.
2Measurement precision
If conventional ultrasonic sensor systems are used, then the resolution of surroundings is maintained, but the detection of poorly reflective objects and objects in motion becomes unreliable
Solution Approach 1:
The patent introduces temporal dimension by analyzing echo signals across multiple time points and using signal processing in the frequency domain. By examining echoes across time and frequency dimensions rather than only spatial resolution, the system reliably detects poorly reflective and moving objects while preserving spatial resolution capabilities.
Solution Approach 2:
The system dynamically adapts its analysis based on detected motion characteristics. By applying different evaluation methods depending on whether objects are static or moving (as determined by distance changes between echoes), the system maintains high detection reliability for both poorly reflective objects and moving objects without compromising resolution.
3Reliability
If echo tracing is performed for dynamic objects, then movement detection capability is improved, but incorrect assignment of echoes to traces increases with changing velocity or surrounding objects
Solution Approach 1:
Instead of maintaining long-term echo traces that require continuous correct assignment, the system uses short-term echo analysis focused on distance changes between consecutive measurements. This disposable approach to echo utilization avoids the accumulation of errors from incorrect trace assignment while still enabling reliable dynamic object identification.
Solution Approach 2:
The patent replaces the mechanical echo tracing system with a field-based analysis of echo signal properties (distance, velocity, acceleration derived from time-series echoes). This substitution eliminates the need for maintaining individual traces and their assignments, thereby preventing incorrect classification while preserving dynamic object detection capability.
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 accuracy and speed of object classification, reducing incorrect classifications and optimizing computing time by focusing on high-quality echo information and monitoring relevant areas around the vehicle.
Implementation Method 1
emitting ultrasonic signals at a plurality of successive time points by means of the ultrasonic sensor system, and receiving ultrasonic echo signals, in particular ultrasonic echo signals reflected on the one or more objects in the surroundings of the vehicle, by means of the ultrasonic sensor system
Data Source
AI summary
A method for classifying at least one object in the surroundings of a vehicle using an ultrasonic sensor system. The method includes: emitting ultrasonic signals at a plurality of successive time points using the ultrasonic sensor system; and receiving ultrasonic echo signals reflected on the one or more objects in the surroundings of the vehicle, using the ultrasonic sensor system; wherein one or more objects in the surroundings are ascertained as a function of the received ultrasonic echo signals, and wherein it is determined as a function of the received ultrasonic echo signals whether the respective object is classified as dynamic, in particular as a pedestrian. A controller for operating an ultrasonic sensor system, an ultrasonic sensor system, and a vehicle including such an ultrasonic sensor system are also described.