Control system of internal combustion engine, electronic control unit, server, and control method of internal combustion engine

Patent metadata

Field	Value
Publication number	US-10825267-B2
Application number	US-201916591968-A
Country	US
Kind code	B2
Filing date	Oct 3, 2019
Priority date	Nov 1, 2018
Publication date	Nov 3, 2020
Grant date	Nov 3, 2020

Abstract

First ECUs and second ECU of a control system of an internal combustion engine is configured to: output a predicted value of an output parameter by using a learning model if actually measured values of input parameters are input; and control an internal combustion engine based on this predicted value, the first ECUs is configured to: learn a learning model; and transmit first vehicle information including a usage environment and usage state of the first vehicle and the learned learning model linked with each other, and the second ECU is configured to receive the learned learning models, and wherein the second ECU is configured to use a learned learning model linked with the first vehicle information closest to the usage environment and usage state of the second vehicle.

First claim

The invention claimed is:

1. A control system of an internal combustion engine comprising first electronic control units, which have a learning function of a learning model used for controlling the internal combustion engine, mounted in a plurality of first vehicles and a second electronic control unit, which does not have the learning function, mounted in a second vehicle,
each of the first electronic control units and the second electronic control unit configured to:
output a predicted value of an output parameter by using a learning model if actually measured values of input parameters are input; and
control the internal combustion engine based on the output predicted value,
the first electronic control units configured to:
learn a learning model used at the first electronic control unit by using teacher data including actually measured values of the input parameters and an actually measured value of the output parameter; and
transmit first vehicle information including at least one of a usage environment and usage state of the first vehicle and the learned learning model linked with each other, and
the second electronic control unit configured to receive the learned learning model, and
wherein the second electronic control unit is configured to use a learned learning model, linked with the first vehicle information in which at least one of the usage environment and usage state of the first vehicle is closest to at least one of the usage environment and usage state of the second vehicle, among the learned learning models learned at the first electronic control units.

2. The control system according to claim 1 , wherein the control system further comprises a server configured to be able to communicate with the first electronic control units and the second electronic control unit,
the second electronic control unit is configured to transmit to the server a model request requesting the learned learning model and second vehicle information including at least one of the usage environment and usage state of the second vehicle, and
the server is configured to:
select the first vehicle information in which at least one of the usage environment and usage state of the first vehicle is closest to at least one of the usage environment and usage state of the second vehicle included in the second vehicle information received from the second electronic control unit, among the first vehicle information received from the first electronic control units; and
transmit the learned learning model linked with the selected first vehicle information to the second electronic control unit.

3. The control system according to claim 1 , wherein
the second electronic control unit is configured to receive the first vehicle information from the first electronic control units,
the second electronic control unit is configured to select the closest first vehicle information, among first vehicle information received from the first electronic control units, and
the second electronic control unit is configured to use the learned learning model linked with the selected first vehicle information.

4. An electronic control unit used in a second vehicle of a control system of an internal combustion engine comprising electronic control units, which have a learning function of a learning model used for controlling the internal combustion engine, mounted in a plurality of first vehicles and an electronic control unit, which does not have the learning function, mounted in the second vehicle,
the electronic control unit of the second vehicle configured to:
output a predicted value of an output parameter by using a learning model if actually measured values of input parameters are input;
control the internal combustion engine of the second vehicle based on the output predicted value output; and
receive the learned learning models learned at the electronic control units of the plurality of first vehicles, and
wherein the electronic control unit of the second vehicle is configured to use said learned learning model learned at the electronic control unit of the first vehicle in which at least one of a usage environment and usage state is closest to at least one of a usage environment and usage state of the second vehicle, among the learned learning models learned at the electronic control units of the plurality of first vehicles.

5. A server configured to be able to communicate with first electronic control units, which have a learning function of a learning model used for controlling an internal combustion engine, mounted in a plurality of first vehicles and a second electronic control unit, which does not have the learning function, mounted in a second vehicle,
the server is configured to:
receive from the first electronic control units first vehicle information including at least one of usage environments and usage states of the first vehicles and learned learning models learned at the first electronic control units and linked with the first vehicle information;
receive from the second electronic control unit a model request requesting the learned learning model and second vehicle information including at least one of a usage environment and usage state of the second vehicle;
select the first vehicle information in which at least one of the usage environment and usage state of the first vehicle is closest to at least one of the usage environment and usage state of the second vehicle included in the received second vehicle information, among the received first vehicle information; and
transmit the learned learning model linked with the selected first vehicle information to the second electronic control unit.

6. A control method of an internal combustion engine in a control system of the internal combustion engine comprising first electronic control units which have a learning function of a learning model used for controlling the internal combustion engine, mounted in a plurality of first vehicles and a second electronic control unit, which does not have the learning function, mounted in a second vehicle,
the control method comprising steps of:
learning, by the first electronic control unit, the learning model used at the first electronic control unit by using teacher data including actually measured values of input parameters of the learning model and actually measured values of output parameters of the learning model;
transmitting, by the first electronic control unit, first vehicle information including at least one of usage environment and usage state of the first vehicle and the learned learning model linked with each other;
receiving, by the second electronic control unit, the learned learning model;
outputting, by the second electronic control unit, a predicted value of an output parameter by using the received learned learning model if actually measured values of input parameters are input; and
controlling, by the second electronic control unit, the internal combustion engine based on the output predicted value,
wherein the step of outputting the predicted value uses the learned learning model, linked with the first vehicle information in which at least one of the usage environment and usage state of the first vehicle is closest to at least one of the usage environment and usage state of the second vehicle, among the learned learning models learned at the first electronic control units.

Assignees

• Toyota Motor Co Ltd

Inventors

• Kitagawa Eiki
• Yokoyama Daiki

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title

• F02D41/2451 Primary

  characterised by what is learned or calibrated · CPC title