Position determination using ultrasound with the aid of an artificial neural network

DE102017123388B4Active Publication Date: 2026-07-23VALEO SCHALTER & SENSOREN GMBH

View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office: DE · DE
Patent Type: Patents
Current Assignee / Owner: VALEO SCHALTER & SENSOREN GMBH
Filing Date: 2017-10-09
Publication Date: 2026-07-23

AI Technical Summary

Technical Problem

Existing systems using artificial neural networks for obstacle detection in motor vehicles require complex training and cannot adapt to different sensor configurations, making them impractical for use in various automobile models with differently located sensors.

Method used

A method utilizing a modular artificial neural network with sensor neuron groups assigned to individual ultrasonic sensors, processing data separately before combining it in a mixer neuron group to determine obstacle coordinates, allowing adaptation to different sensor arrangements without additional training.

Benefits of technology

Enables efficient position determination of obstacles using ultrasound by reducing the need for retraining the neural network when sensor configurations change, facilitating its use across different vehicle models.

✦ Generated by Eureka AI based on patent content.

Smart Images

Figure 00000000_0000_ABST

Patent Text Reader

Abstract

Method for determining the position of an obstacle (2) using ultrasound by means of a plurality of ultrasonic sensors (21, 22, 23, 24, 25, 26) and an artificial neural network (5) connected to the ultrasonic sensors (21, 22, 23, 24, 25, 26), wherein the ultrasonic sensors (21, 22, 23, 24, 25, 26) are arranged side by side at predetermined distances from each other, the artificial neural network (5) comprises a plurality of sensor neuron groups (31, 32, 33, 34, 35, 36) and a mixer neuron group (40), each ultrasonic sensor (21, 22, 23, 24, 25, 26) is assigned a sensor neuron group (31, 32, 33, 34, 35, 36) with which only data from its assigned ultrasonic sensor (21, 22, 23, 24, 25, 26), and all sensor neuron groups (31, 32, 33, 34, 35, 36) are connected to the mixer neuron group (40), so that the mixer neuron group (40) receives data from all sensor neuron groups (31, 32, 33, 34, 35, 36),with the following process steps: Emitting at least one ultrasound signal, Receiving ultrasound signal echoes at at least one ultrasound sensor (21, 22, 23, 24, 25, 26), Determining distance values (e1, e2, e3, e4) for the received ultrasound signal echoes based on the transit times of the ultrasound signal echoes, Inputting the distance values (e1, e2, e3, e4) into the sensor neuron groups (31, 32, 33, 34, 35, 36), Processing the distance values (e1, e2, e3, e4) in the sensor neuron groups (31, 32, 33, 34, 35, 36), Outputting the distance values processed in the sensor neuron groups (31, 32, 33, 34, 35, 36) from the sensor neuron groups (31, 32, 33, 34, 35, 36) and inputting the distance values processed in the sensor neuron groups (31, 32, 33, 34, 35, 36) into the mixer neuron group (40), processing the distance values processed in the sensor neuron groups (31, 32, 33, 34, 35, 36) in the mixer neuron group (40) and outputting at least two-dimensional coordinates (X',Y') for the obstacle (2) detected by the emitted ultrasound signal by the mixer neuron group (40).,

Need to check novelty before this filing date? Find Prior Art