Ultrasonic Object Classification Using Reflection Point Dispersion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional ultrasonic-based object classification methods in vehicles suffer from high error rates due to insufficient utilization of parameters for distinguishing relevant objects like pedestrians, walls, or trees from irrelevant objects like curbs.
Innovation Solution
The method employs ultrasonic sensors with overlapping fields of view to determine the position of reflection points using lateration, and utilizes dispersion parameters of these points, such as mean value, standard deviation, and variance, to classify objects into categories like 'low, crossable' or 'high, non-crossable', and further differentiates between types like curbs, pedestrians, and posts using additional criteria like echo behavior and amplitudes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional ultrasonic-based object classification methods are used, then the system can detect objects in the environment, but the classification error rate is high due to insufficient utilization of parameters
Solution Approach 1:
The patent applies parameter changes by utilizing multiple parameters including dispersion parameters (standard deviation, variance) of reflection point positions, echo behavior characteristics, and amplitude variations. This transforms the classification system from using limited parameters to comprehensively analyzing multiple parameters, thereby resolving the contradiction between reliability and information loss.
2Measurement precision
If multiple ultrasonic sensors with overlapping fields of view are used to determine reflection point positions, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the environment into multiple detection zones corresponding to different ultrasonic sensors. Each sensor independently measures reflection points in its field of view, and the control unit segments the analysis by determining dispersion parameters separately for each object based on reflected signals from different sensors. This resolves the contradiction by enabling precise multi-sensor measurement while maintaining manageable system complexity through segmented processing.
3Reliability
If dispersion parameters and echo characteristics are analyzed to differentiate object types, then object classification accuracy improves, but computational requirements and processing time increase
Solution Approach 1:
The patent applies preliminary action by continuously determining dispersion parameters and allocating reflection points to objects in advance, before critical classification decisions are needed. The control unit maintains updated information about object characteristics including dispersion parameters and echo behaviors, enabling rapid classification when needed without intensive real-time computation, thus resolving the contradiction between accuracy and processing time.
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 significantly reduces error rates, particularly improving the differentiation between curbs and pedestrians, and enhances the true-positive detection of trees and bushes by considering dispersion parameters and echo characteristics.
Implementation Method 1
ultrasonic signals are emitted with the aid of ultrasonic sensors, ultrasonic echoes are received from objects in the environment
Implementation Method 2
ultrasonic echoes are received from objects in the environment, and the position of a reflection point relative to the ultrasonic sensors is determined
Implementation Method 3
the position of a reflection point relative to the ultrasonic sensors is determined with the aid of lateration
Implementation Method 4
From the propagation time of the ultrasonic signals up to the point when a corresponding ultrasonic echo is received and also the known speed of sound, the distance between a reflecting object and the respective sensor is ascertainable
Data Source
AI summary
A method for classifying objects in the environment of a vehicle with the aid of ultrasonic sensors. In the method, ultrasonic signals are emitted, ultrasonic echoes are received from objects in the environment, and the position of a reflection point relative to the ultrasonic sensors is determined using lateration, reflection points being continuously determined and the reflection points being allocated to objects in the environment. Dispersion parameters relating to the position of the reflection points allocated to an object are determined and used as a classification criterion with regard to the type of object. A driver assistance system and a vehicle including such a driver assistance system, are also described.