DRIVER STATE ASSESSMENT DEVICE

• US 20120221171A1
• Filed: 09/28/2010
• Published: 08/30/2012
• Est. Priority Date: 09/30/2009
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Sugimoto, Yoichi, Shirakata, Tetsuro, Tokutake, Hiroshi
• Granted Patent: US 8,489,253 B2
• Time in Patent Office: Days
• US Class Current: 701/1
• CPC Class Codes: B60W 2040/0827 due to sleepiness
• CPC Class Codes: B60W 2040/0836 due to alcohol
• CPC Class Codes: B60W 2050/0029 Mathematical model of the driver
• CPC Class Codes: B60W 2420/403 Image sensing, e.g. optical camera
• CPC Class Codes: B60W 2520/14 Yaw
• CPC Class Codes: B60W 2540/18 Steering angle
• CPC Class Codes: B60W 40/08 related to drivers or passengers
• CPC Class Codes: G08B 21/06 indicating a condition of sleep, e.g. anti-dozing alarms

Abstract

A driver state assessment device is provided in which when driver model identification means (M2) identifies a driver model showing a driver'"'"'s input/output relationship using a difference between a target azimuth and an actual azimuth as a driver'"'"'s input and an actual steering angle as a driver'"'"'s output, driver model amount of operation acquisition means (M3) acquires a driver model steering angle by inputting a current azimuth deviation into the driver model, and driver state assessment means (M4) calculates a difference between a current actual steering angle and a driver model steering angle as a residual error and assesses the driver'"'"'s state based on the residual error. Therefore, since the residual error is an index that represents a fluctuation component, a noise component, a non-linear component, etc. obtained from the driver model, it is possible to assess, with high precision, the driver'"'"'s state, in particular a low wakefulness state of the driver, based on this residual error.

Claims

1. A driver state assessment device comprising

operating target value acquisition device for determining a driver'"'"'s operating target value based on an output from a vehicle-installed sensor,

actual amount of movement acquisition device for determining an actual amount of movement of the vehicle,

actual amount of operation acquisition device for determining an actual amount of operation by the driver,

driver model identification device for identifying a driver model showing a driver'"'"'s input/output relationship using a difference between the operating target value and the actual amount of movement as the driver'"'"'s input and the actual amount of operation as the driver'"'"'s output,

driver model amount of operation acquisition device for acquiring a driver model amount of operation by inputting a difference between a current operating target value and the actual amount of movement into the driver model identified by the driver model identification device, and

driver state assessment device for calculating a difference between the current actual amount of operation determined by the actual amount of operation acquisition device and the driver model amount of operation determined by the driver model amount of operation acquisition device as a residual error, and assessing the driver'"'"'s state based on the residual error.

2. The driver state assessment device according to claim 1, wherein the driver state assessment device calculates a normalized residual error from smoothed residual error/(steady state gain)², and assesses the driver'"'"'s state based on the normalized residual error.

3. The driver state assessment device according to claim 2, wherein the driver state assessment device calculates the normalized residual error based on the value of the steady state gain, for which the frequency is 10 rad/sec.

4. The driver state assessment device according to claim 2, wherein the driver state assessment device assesses that the driver is in a low wakefulness state when the normalized residual error is at least an assessment threshold value.

5. The driver state assessment device according to claim 1, wherein the driver model identification device identifies a driver model by a first-order differential expression.

6. The driver state assessment device according to claim 1, wherein the device comprises a filter device for restricting the frequency range of a signal inputted into the driver model.

7. The driver state assessment device according to claim 6, wherein the filter device restricts the frequency range of the signal inputted into the driver model so as to be from 1 rad/sec to 10 rad/sec.

8. The driver state assessment device according to claim 3, wherein the driver state assessment device assesses that the driver is in a low wakefulness state when the normalized residual error is at least an assessment threshold value.