Understanding driver awareness through brake behavior analysis

• US 10,279,793 B2
• Filed: 05/11/2017
• Issued: 05/07/2019
• Est. Priority Date: 05/11/2017
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Li, Nanxiang, Misu, Teruhisa, Tao, Fei
• Primary Examiner(s): Kleinman, Lail A
• Application Number:
  US15/593,087
• Publication Number:
  US 20180326966A1
• Time in Patent Office: 726 Days
• Field of Search: None
• CPC Class Codes: B60T 17/22 Devices for monitoring or checking brake systems; Signal devices
• CPC Class Codes: B60T 2220/02 Driver type; Driving style; Driver adaptive features
• CPC Class Codes: B60T 7/14 operated upon collapse of driver deadman's devices for electrically propelled vehicles B60L3/02
• CPC Class Codes: B60W 2040/0872 Driver physiology
• CPC Class Codes: B60W 2050/0002 Automatic control, details of type of controller or control system architecture
• CPC Class Codes: B60W 2420/408 Radar; Laser, e.g. lidar
• CPC Class Codes: B60W 2520/105 Longitudinal acceleration
• CPC Class Codes: B60W 2540/22 Psychological state; Stress level or workload
• CPC Class Codes: B60W 2540/221 Physiology, e.g. weight, heartbeat, health or special needs
• CPC Class Codes: B60W 2554/00 Input parameters relating to objects
• CPC Class Codes: B60W 2556/10 Historical data
• CPC Class Codes: B60W 30/18109 Braking
• CPC Class Codes: B60W 40/09 Driving style or behaviour

Abstract

Systems and methods for analyzing a driver'"'"'s brake behavior are provided. In some aspects, the system includes a memory that stores instructions for executing processes for analyzing the driver'"'"'s brake behavior, and a processor configured to execute the instructions. In aspects, the processes include: receiving data from one or more multi-modal sensors, determining whether each braking event of a plurality of braking events is a hard braking event; analyzing the one or more physiological signals of the driver to determine whether each of the hard braking events is a reactive hard braking event or an intended hard braking event; applying a Lasso regression analysis and a classification framework for each of the hard braking events to identify features that indicate whether a respective hard braking event is a reactive hard braking event or an intended hard braking event; and generating a signal notifying a vehicle of the features.

Claims

1. A system for analyzing a driver'"'"'s brake behavior, the system comprising:

a memory that stores instructions for executing processes for analyzing the driver'"'"'s brake behavior; and

a processor configured to execute the instructions, wherein the processes comprise;

receiving data from one or more multi-modal sensors, the data including one or more characteristics of a road and one or more physiological signals of the driver;

determining whether each braking event of a plurality of braking events is a hard braking event;

analyzing the one or more physiological signals of the driver to determine whether each of the hard braking events is a reactive hard braking event or an intended hard braking event, wherein the analyzing comprises determining whether a change of each of the one or more physiological signals of the driver exceeds a threshold value during the hard braking event;

applying a Lasso regression analysis and a classification framework for each of the hard braking events to identify road features that indicate whether a respective hard braking event is a reactive hard braking event or an intended hard braking event; and

generating a signal notifying a vehicle of the features that indicate whether the respective hard braking event is a reactive hard braking event or an intended hard braking event.

2. The system of claim 1, wherein, to determine whether a braking event is a hard braking event, the processor is further configured to determine whether an intensity of the braking event exceeds a threshold value, and wherein a braking event exceeding the threshold value is a hard braking event.

3. The system of claim 2, wherein the intensity of the braking event is based on a maximum deceleration during the braking event.

4. The system of claim 1, wherein:

when the one or more physiological signals exceeds the threshold value, the hard braking event is a reactive hard braking event; and

when the one or more physiological signals remains below the threshold value, the hard braking event is determined to be an intended hard braking event.

5. The system of claim 1, wherein the one or more multi-modal sensors includes a Lidar scanner, and wherein, to apply the Lasso regression analysis, the processor is further configured to:

down-sample a plurality of signals received from the Lidar scanner;

determine a plurality statistics for each of the down-sampled signals to generate a plurality of dimensional vectors; and

apply the Lasso regression based on the dimensional vectors and the determination of whether each of the hard braking events is a reactive hard braking event or an intended hard braking event.

6. The system of claim 5, wherein, to apply the classification framework, the processor is further configured to apply a binary classification to identify Lidar features that distinguish between reactive hard braking events and intended hard braking events.

7. A method for analyzing a driver'"'"'s brake behavior, the method comprising:

receiving data from one or more multi-modal sensors, the data including one or more characteristics of a road and one or more physiological signals of the driver;

determining whether each braking event of a plurality of braking events is a hard braking event;

analyzing the one or more physiological signals of the driver to determine whether each of the hard braking events is a reactive hard braking event or an intended hard braking event, wherein analyzing the one or more physiological signals comprises determining whether a change of each of the one or more physiological signals of the driver exceeds a threshold value during the hard braking event;

applying a Lasso regression analysis and a classification framework for each of the hard braking events to identify features that indicate whether a respective hard braking event is a reactive hard braking event or an intended hard braking event; and

generating a signal notifying a vehicle of the features that that indicate whether a respective hard braking event is a reactive hard braking event or an intended hard braking event.

8. The method of claim 7, wherein determining whether a braking event is a hard braking event comprises determining whether an intensity of the braking event exceeds a threshold value, and wherein a braking event exceeding the threshold value is a hard braking event.

9. The method of claim 8, wherein the intensity of the braking event is based on a maximum deceleration during the braking event.

10. The method of claim 7, wherein:

when the one or more physiological signals exceeds the threshold value, the hard braking event is a reactive hard braking event; and

when the one or more physiological signals remains below the threshold value, the hard braking event is determined to be an intended hard braking event.

11. The method of claim 7, wherein the one or more multi-modal sensors includes a Lidar scanner, and wherein, applying the Lasso regression analysis comprises:

down-sampling a plurality of signals received from the Lidar scanner;

determining a plurality statistics for each of the down-sampled signals to generate a plurality of dimensional vectors; and

applying the Lasso regression based on the dimensional vectors and the determination of whether each of the hard braking events is a reactive hard braking event or an intended hard braking event.

12. The method of claim 11, wherein applying the classification framework comprises applying a binary classification to identify Lidar features that distinguish between reactive hard braking events and intended hard braking events.

13. A non-transitory computer-readable storage medium containing executable computer program code, the code comprising instructions configured to:

receive data from one or more multi-modal sensors, the data including one or more characteristics of a road and one or more physiological signals of the driver;

determine whether each braking event of a plurality of braking events is a hard braking event;

analyze the one or more physiological signals of the driver to determine whether each of the hard braking events is a reactive hard braking event or an intended hard braking event, wherein analyzing the one or more physiological signals comprises determining whether a change of each of the one or more physiological signals of the driver exceeds a threshold value during the hard braking event;

apply a Lasso regression analysis and a classification framework for each of the hard braking events to identify features that indicate whether a respective hard braking event is a reactive hard braking event or an intended hard braking event; and

generate a signal notifying a vehicle of the features that that indicate whether a respective hard braking event is a reactive hard braking event or an intended hard braking event.

14. The medium of claim 13, wherein, to determine whether a braking event is a hard braking event, the code further comprises instructions configured to determine whether an intensity of the braking event exceeds a threshold value, and wherein a braking event exceeding the threshold value is a hard braking event.

15. The medium of claim 14, wherein the intensity of the braking event is based on a maximum deceleration during the braking event.

16. The medium of claim 13, wherein:

when the one or more physiological signals exceeds the threshold value, the hard braking event is a reactive hard braking event; and

when the one or more physiological signals remains below the threshold value, the hard braking event is determined to be an intended hard braking event.

17. The medium of claim 13, wherein the one or more multi-modal sensors includes a Lidar scanner, and wherein:

to apply the Lasso regression analysis, the code further comprises instructions configured to;

down-sample a plurality of signals received from the Lidar scanner,

determine a plurality statistics for each of the down-sampled signals to generate a plurality of dimensional vectors, and

apply the Lasso regression based on the dimensional vectors and the determination of whether each of the hard braking events is a reactive hard braking event or an intended hard braking event; and

to apply the classification framework, the code further comprises instructions configured to;

apply a binary classification to identify Lidar features that distinguish between reactive hard braking events and intended hard braking events.