Granted Patent B2

US 12,525,111 · App. 17/231,836 · Granted Jan 13, 2026

Inventors: Poorna Kale (Folsom, CA); Robert Richard Noel Bielby (Placerville, CA)

G08B21/06A61B5/18G06N20/00G06V10/764G06V10/7784G06V10/82G06V20/597B60K28/06G06N3/02

Abstract

Claims (47)

1 . A vehicle comprising:

a camera configured to generate images;

an artificial neural network configured to analyze the images, wherein the artificial neural network is trained using training images captured from persons; and

a computer system configured to provide instructions in response to a classification by the artificial neural network, wherein the instructions are configured to request a driver to provide a response;

wherein the computer system is further configured to:

store the images from the camera in an input namespace, and store output data from the artificial neural network in an output namespace; and

configure memory operations to write the images in the input namespace at a first frequency, and write the output data in the output namespace at a second frequency, wherein the first frequency is higher than the second frequency.

2 . The vehicle of claim 1 , wherein the response is detectable in subsequent images captured by the camera.

3 . The vehicle of claim 1 , wherein the driver can provide the response by moving at least one of a viewing direction of the driver's eyes, a hand of the driver, the driver's head, or one or more fingers of the driver.

4 . The vehicle of claim 1 , wherein the classification is confirmed based on lack of the response requested via the instructions.

5 . The vehicle of claim 1 , further comprising an infotainment system, wherein the computer system provides the instructions via the infotainment system.

6 . The vehicle of claim 1 , wherein the instructions are provided at varying volume levels.

7 . The vehicle of claim 1 , wherein the computer system is implemented with a data storage device, and the data storage device is configured to:

store model data of the artificial neural network; and

receive, from the camera, the images.

8 . The vehicle of claim 1 , wherein the camera is configured on a rear view mirror of the vehicle, a side view mirror of the vehicle, and/or an instrument panel of the vehicle.

9 . The vehicle of claim 1 wherein the classification is confirmed by analyzing subsequent images captured by the camera.

10 . The vehicle of claim 1 , wherein the images are captured during time periods in which the persons are in a first state of alertness, and during other time periods in which the persons are in a second state of alertness.

11 . The vehicle of claim 1 , wherein the instructions request a physical movement of the driver.

12 . The vehicle of claim 1 , wherein the computer system is further configured to store model data of the artificial neural network in a model namespace, and to configure the memory operations in the model namespace according to a data usage pattern of the artificial neural network.

13 . The vehicle of claim 12 , wherein the computer system is further configured to use an address map having namespace settings to configure the memory operations in the input namespace, the model namespace, and the output namespace.

14 . An apparatus comprising:

a sensor configured to generate first images; and

a data storage device configured to store model data of an artificial neural network, wherein in response to receiving the first images in the data storage device, the data storage device is configured to analyze the first images;

wherein the apparatus is configured to communicate, to a computer system of a vehicle, a classification by the artificial neural network;

wherein communicating the classification causes the vehicle to perform an action;

wherein the data storage device is further configured to use an address map having namespace settings to configure memory operations in an input namespace for the first images, a model namespace for model data of the artificial neural network, and an output namespace for the classification; and

wherein the namespace settings indicate different access characteristics for each of the input namespace, the model namespace, and the output namespace.

15 . The apparatus of claim 14 , wherein:

the sensor is further configured to generate subsequent images following the action; and

the data storage device is further configured to determine a confirmation of the classification based on an analysis of the subsequent images.

16 . The apparatus of claim 14 , wherein analyzing the first images comprises extracting features from the first images so that an orientation of the head of a driver can be measured from the first images.

17 . The apparatus of claim 14 , wherein the artificial neural network is trained using images of a driver during time periods in which the driver is not drowsy.

18 . The apparatus of claim 14 , wherein the data storage device comprises a buffer configured to store the model data and an output from the artificial neural network, and wherein the data storage device is further configured to store the output from the artificial neural network in the output namespace.

19 . A method comprising:

storing model data of an artificial neural network in a model namespace of a data storage device;

storing input data for the artificial neural network in an input namespace;

configuring memory operations in the model namespace according to a data usage pattern of the artificial neural network;

configuring memory operations in the input namespace according to a data usage pattern of the input data;

generating a sequence of images;

analyzing, by the artificial neural network, the sequence of images to provide a classification, wherein the artificial neural network is trained using images captured from a driver of a vehicle; and

in response to the classification, requesting the driver provide a response that includes a physical movement.

20 . The method of claim 19 , further comprising in response to the classification, validating, by an advanced driver assistance system, control signals provided by the driver.

21 . The method of claim 19 , further comprising training the artificial neural network to recognize patterns in training images generated by a camera of the vehicle, wherein the patterns are associated with the driver being in an alert state, and the training images are generated during a time period in which the driver is in the alert state.

22 . The method of claim 19 , further comprising confirming the classification based on sensor data collected by the vehicle after requesting the response.

23 . The method of claim 19 , further comprising rejecting the classification based on sensor data collected by the vehicle after requesting the response.

24 . The method of claim 19 , wherein the artificial neural network is trained by the driver making the physical movement.

Assignments (2)

Continuity (2)

References Cited (152)