US 2026/0145693 A1 · DEVICE AND METHOD FOR DIAGNOSING FAULTS IN ELECTRONIC BRAKE SYSTEM BASED ON DEEP LEARNING MODEL

Abstract

According to an embodiment of the present disclosure, a device and a method for diagnosing faults in an electronic brake system based on a deep learning model, which may be provided, which collect and preprocess vehicle state data including sensor data and actuator state data through an in-vehicle communication network, input the preprocessed vehicle state data into a deep learning anomaly detection model to determine a normal state or a fault state of the electronic brake system, input the preprocessed vehicle state data into a deep learning detailed fault diagnosis model to determine a fault type and fault severity of the electronic brake system, if the electronic brake system is determined to be in the fault state, and independently control the electronic brake system based on the fault type and fault severity.

Claims20

Dependent claims

A device for diagnosing faults in an electronic brake system based on a deep learning model, comprising:

a signal preprocessor collecting and preprocessing vehicle state data including sensor data and actuator state data through an in-vehicle communication network;
a system anomaly detector inputting the preprocessed vehicle state data into a deep learning anomaly detection model to determine a normal state or a fault state of the electronic brake system;
a detailed fault diagnoser inputting the preprocessed vehicle state data into a deep learning detailed fault diagnosis model to determine a fault type and fault severity of the electronic brake system, if the electronic brake system is determined to be in the fault state;
and
a system controller independently controlling the electronic brake system based on the fault type and fault severity.

2 more elements

A device for diagnosing faults in an electronic brake system based on a deep learning model, comprising:

at least one memory including computer program instructions and at least one processor executing the computer program instructions,
wherein the at least one processor collects and preprocesses vehicle state data including sensor data and actuator state data through an in-vehicle communication network, inputs the preprocessed vehicle state data into a deep learning anomaly detection model to determine a normal state or a fault state of the electronic brake system, inputs the preprocessed vehicle state data into a deep learning detailed fault diagnosis model to determine a fault type and fault severity of the electronic brake system, if the electronic brake system is determined to be in the fault state, and independently controls the electronic brake system based on the fault type and fault severity.

Dependencies are inferred from the claim text.

Device and method for diagnosing faults in electronic brake system based on deep learning model

Life of this patent

Device and method for diagnosing faults in electronic brake system based on deep learning model

Life of this patent

Abstract

Claims20