Method and system of assisting driving of vehicle

• US 10,668,929 B2
• Filed: 02/26/2018
• Issued: 06/02/2020
• Est. Priority Date: 03/29/2017
• Status: Active Grant

• Current Assignee: Mazda Motor Corporation
• Original Assignee: Mazda Motor Corporation
• Inventors: Tochioka, Takahiro, Goto, Seiji, Okano, Hideki, Ikeda, Kazuhiro
• Primary Examiner(s): Shaawat, Mussa A
• Application Number:
  US15/904,851
• Publication Number:
  US 20180281812A1
• Time in Patent Office: 827 Days
• CPC Class Codes: B60W 2050/0029 Mathematical model of the driver
• CPC Class Codes: B60W 2050/0088 Adaptive recalibration
• CPC Class Codes: B60W 2540/043 Identity of occupants
• CPC Class Codes: B60W 2540/22 Psychological state; Stress level or workload
• CPC Class Codes: B60W 30/025 related to comfort of drivers or passengers
• CPC Class Codes: B60W 40/09 Driving style or behaviour
• CPC Class Codes: B60W 50/0098 Details of control systems ensuring comfort, safety or stability not otherwise provided for

Abstract

A vehicle drive assistance system is provided, which includes one or more processors configured to execute a general driver model learning engine configured to build a general driver model to be applied to a plurality of drivers based on driving data of the drivers, an individual driver model learning engine configured to build an individual driver model unique to a specific driver based on driving data of the driver, and an on-board controller provided in a vehicle operated by the driver. The individual driver model learning engine includes a vehicle control updating program configured to cause the on-board controller to update vehicle control processing based on the general and individual driver models. The vehicle control updating program acquires the driver models and, according to a given condition, determines a driver model based on which the vehicle control processing is updated, between the general and individual driver models.

Claims

1. A vehicle drive assistance system, comprising one or more processors configured to execute:

a general driver model learning engine configured to build a general driver model to be applied to a plurality of vehicle drivers based on driving data of the plurality of drivers; and

an individual driver model learning engine configured to build an individual driver model unique to a specific vehicle driver based on driving data of the specific driver; and

an on-board controller provided in a vehicle operated by the specific driver and configured to perform a particular vehicle control processing,

wherein the individual driver model learning engine includes a vehicle control updating program configured to cause the on-board controller to update the vehicle control processing based on the general driver model and the individual driver model,

wherein the vehicle control updating program acquires the general driver model and the individual driver model and, according to a given condition, determines a driver model based on which the vehicle control processing is updated, between the general driver model and the individual driver model,

wherein the individual driver model learning engine includes a driver state analyzing program configured to analyze a current state of the specific driver based on the driving data of the specific driver, and

wherein when the driver state analyzed by the driver state analyzing program is not changed to a given state after the vehicle control processing is updated based on the individual driver model, the vehicle control updating program updates the vehicle control processing based on the general driver model.

2. The vehicle drive assistance system of claim 1, wherein the vehicle control updating program determines a difference between the general driver model and the individual driver model, and when the difference is above a given threshold, the vehicle control processing is updated based on the general driver model.

3. The vehicle drive assistance system of claim 1, wherein the vehicle control updating program prioritizes the update of the vehicle control processing based on the individual driver model than the update of the vehicle control processing based on the general driver model.

4. A method of assisting driving of a vehicle by a vehicle drive assistance system including one or more processors configured to execute a general driver model learning engine configured to build a general driver model to be applied for a plurality of vehicle drivers based on driving data of the plurality of drivers, and an individual driver model learning engine configured to learn an individual driver model unique to a specific vehicle driver based on driving data of the specific driver, and an on-board controller provided in a vehicle operated by the specific driver and configured to program particular vehicle control processing, the method comprising causing the individual driver model learning engine to:

acquire the general driver model from the general driver model learning engine;

acquire a control parameter of the vehicle control processing from the on-board controller;

acquire from the individual driver model an individual driver model parameter corresponding to the acquired control parameter;

acquire from the general driver model a general driver model parameter corresponding to the acquired control parameter;

calculate a difference between the individual driver model parameter and the general driver model parameter;

calculate an update parameter for the control parameter based on the individual driver model parameter when the difference is smaller than a given value;

calculate an update parameter for the control parameter based on the general driver model parameter when the difference is larger than the given value;

transmit to the on-board controller an instruction for updating the control parameter to the update parameter;

execute a driver state analyzing program to analyze a current state of the specific driver based on the driving data of the specific driver; and

when the driver state analyzed by the driver state analyzing program is not changed to a given state after the vehicle control processing is updated based on the individual driver model, updating the vehicle control processing based on the general driver model.

5. A method of assisting driving of a vehicle by a vehicle drive assistance system including one or more processors configured to execute a general driver model learning engine configured to build a general driver model to be applied for a plurality of vehicle drivers based on driving data of the plurality of drivers, and an individual driver model learning engine configured to build an individual driver model unique to a specific vehicle driver based on driving data of the specific driver, and an on-board controller provided in a vehicle operated by the specific driver and configured to execute a particular vehicle control processing, the method comprising:

causing a driver state analyzing program to analyze a current emotional state of the driver based on the driving data of the specific driver;

causing the individual driver model learning engine to acquire a control parameter of the vehicle control processing from the on-board controller;

causing the individual driver model learning engine to acquire from the individual driver model an individual driver model parameter corresponding to the acquired control parameter;

causing the individual driver model learning engine to calculate an update parameter for the control parameter based on the individual driver model parameter;

causing the individual driver model learning engine to transmit to the on-board controller an instruction for updating the control parameter to the update parameter;

causing the on-board controller to update the control parameter to the update parameter;

causing the driver state analyzing unit to analyze the emotional state of the driver based on the driving data of the specific driver after the update;

causing the individual driver model learning engine to acquire a general driver model parameter corresponding to the updated control parameter from the general driver model when the driver emotional state analyzed by the driver state analyzing program is not changed to a given state after the update;

causing the individual driver model learning engine to calculate an update parameter for the updated control parameter based on the general driver model parameter; and

causing the individual driver model learning engine to transmit to the on-board controller an instruction for updating the updated control parameter to the update parameter calculated based on the general driver model parameter.