IP Library Granted Patent US 12662136
Granted Patent B1
US 12662136 · App. 18/834,747 · Granted Jun 23, 2026

Driving skill evaluation method, driving skill evaluation system, and non-transitory recording medium

Inventors: Takeshi Torii (Tokyo, JP); Noeru Sato (Tokyo, JP)
Assignee: SUBARU CORPORATION
B60W40/09B60W50/14G06V10/751G06V20/588B60W2420/403B60W2520/06B60W2552/30
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12662136
App. No.
18/834,747
Granted
Jun 23, 2026
Kind
B1
Abstract

A driving skill evaluation method according to an embodiment of the disclosure includes: performing a detection process of detecting a curve based on traveling data of a vehicle; and performing an evaluation process of evaluating a driving skill of a driver of the vehicle based on the traveling data at the curve. The evaluation process includes: generating a first kernel density estimation image based on the traveling data at a first curve detected by the detection process; calculating first similarity levels between the first kernel density estimation image and reference images associated with each of curves excluding the first curve, and calculating, for each of the curves, a first average similarity level that is an average value of the first similarity levels; and evaluating the driving skill of the driver based on the first average similarity level indicating a highest degree of similarity, of the first average similarity levels.

Claims (39)

1 . A driving skill evaluation method comprising:

performing a detection process of detecting a curve based on traveling data of a vehicle; and

performing an evaluation process of evaluating a driving skill of a driver of the vehicle based on the traveling data at the curve, wherein

the evaluation process comprises

generating a first kernel density estimation image based on the traveling data at a first curve detected by the detection process,

calculating first similarity levels between the first kernel density estimation image and reference images associated with each of curves excluding the first curve, and calculating, for each of the curves, a first average similarity level that is an average value of the first similarity levels, and

evaluating the driving skill of the driver of the vehicle based on the first average similarity level indicating a highest degree of similarity, of the first average similarity levels.

2 . The driving skill evaluation method according to claim 1 , wherein

the curves include neither the first curve nor a second curve,

the evaluation process further comprises

generating a second kernel density estimation image based on the traveling data at the second curve detected by the detection process, and

calculating second similarity levels between the second kernel density estimation image and the reference images associated with each of the curves, and calculating, for each of the curves, a second average similarity level that is an average value of the second similarity levels, and

when the driving skill of the driver of the vehicle is to be evaluated, the driving skill of the driver of the vehicle is evaluated based on the second average similarity level indicating a highest degree of similarity, of the second average similarity levels, in addition to the first average similarity level.

3 . The driving skill evaluation method according to claim 1 , wherein

the curves are provided in a predetermined area, and

the detection process allows detection of the curve when the curve is present in an evaluation target area different from the predetermined area.

4 . The driving skill evaluation method according to claim 1 , wherein

a first image direction in the kernel density estimation image represents time,

a second image direction in the kernel density estimation image represents a first parameter corresponding to a direction change in a traveling direction of the vehicle, and

a pixel value of the kernel density estimation image corresponds to data of a second parameter indicating a square of a jerk in the traveling direction of the vehicle.

5 . The driving skill evaluation method according to claim 1 , wherein

a first image direction in the kernel density estimation image represents time,

a second image direction in the kernel density estimation image represents a first parameter indicating a square of a jerk in a traveling direction of the vehicle, and

a pixel value of the kernel density estimation image corresponds to data of a second parameter corresponding to a direction change in the traveling direction of the vehicle.

6 . The driving skill evaluation method according to claim 1 , further comprising presenting an evaluation result of the driving skill of the driver to the driver.

7 . A driving skill evaluation system comprising:

a curve detection circuit configured to perform a detection process of detecting a curve based on traveling data of a vehicle; and

an evaluation circuit configured to perform an evaluation process of evaluating a driving skill of a driver of the vehicle based on the traveling data at the curve, wherein

the evaluation process comprises

generating a first kernel density estimation image based on the traveling data at a first curve detected by the detection process,

calculating first similarity levels between the first kernel density estimation image and reference images associated with each of curves excluding the first curve, and calculating, for each of the curves, a first average similarity level that is an average value of the first similarity levels, and

evaluating the driving skill of the driver of the vehicle based on the first average similarity level indicating a highest degree of similarity, of the first average similarity levels.

8 . A non-transitory recording medium containing software, the software causing a processor to:

perform a detection process of detecting a curve based on traveling data of a vehicle; and

perform an evaluation process of evaluating a driving skill of a driver of the vehicle based on the traveling data at the curve, wherein

the evaluation process comprises

generating a first kernel density estimation image based on the traveling data at a first curve detected by the detection process,

calculating first similarity levels between the first kernel density estimation image and reference images associated with each of curves excluding the first curve, and calculating, for each of the curves, a first average similarity level that is an average value of the first similarity levels, and

evaluating the driving skill of the driver of the vehicle based on the first average similarity level indicating a highest degree of similarity, of the first average similarity levels.