System and method for monitoring driving behavior of a driver

• US 9,702,703 B2
• Filed: 03/18/2016
• Issued: 07/11/2017
• Est. Priority Date: 03/27/2015
• Status: Active Grant

• Current Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: Chowdhury, Arijit, Chakravarty, Tapas, Banerjee, Tanushree, Purushothaman, Balamuralidhar
• Primary Examiner(s): Black, Thomas G
• Assistant Examiner(s): HUYNH, LUAT T
• Application Number:
  US15/074,641
• Publication Number:
  US 20160282124A1
• Time in Patent Office: 480 Days
• Field of Search: 701 70, 701 324, 701 337
• CPC Class Codes: B60W 40/09 Driving style or behaviour
• CPC Class Codes: G01C 21/165 combined with non-inertial navigation instruments
• CPC Class Codes: G07C 5/0816 Indicating performance data, e.g. occurrence of a malfunction

Abstract

This disclosure relates generally to data processing, and more particularly to a system and method for monitoring driving behavior of a driver. In one embodiment, a system (102) for monitoring driving behavior of a driver is disclosed. The system (102) may configure a processor (202) to execute computer-readable instructions (208) stored in a memory (206) in order to: capture a plurality of acceleration samples; compute Kurtosis values and Skewness values corresponding to a set of acceleration samples; filter the Kurtosis values; determine a probability distribution function of the filtered Kurtosis values; compute a mean and a standard deviation associated with the filtered Kurtosis values; determine a first threshold for each driver based upon the mean and the standard deviation; compute a first score for each driver based upon the first threshold and the number of trips; determine a second threshold; compute a second score for each driver based upon the second threshold and the number of trips; and evaluate driving behavior of a driver based upon the first score and the second score.

Claims

1. A system for monitoring driving behavior of a driver, the system comprising:

a processor; and

a memory coupled to the processor, wherein the processor is configured to execute computer-readable instructions stored in the memory in order to;

capture a plurality of acceleration samples associated with a plurality of trips completed by a plurality of drivers at a predefined time interval, wherein the plurality of acceleration samples is segregated into multiple windows such that each window comprises acceleration samples captured corresponding to a predetermined time window for each driver;

compute Kurtosis values and Skewness values corresponding to a set of acceleration samples, of the plurality of acceleration samples, associated with each driver;

filter the Kurtosis values to obtain filtered Kurtosis values, wherein the Kurtosis values are filtered based upon the Skewness values;

determine a probability distribution function of the filtered Kurtosis values;

compute a mean and a standard deviation associated with the filtered Kurtosis values based upon the probability distribution function;

determine a first threshold for each driver based upon the mean and the standard deviation;

compute a first score for each driver based upon a number of trips having a Kurtosis value greater than the first threshold and the plurality of the trips;

determine a second threshold, wherein the second threshold is a minimum of the first threshold determined for each of the plurality of drivers;

compute a second score for each driver based upon a number of trips having a Kurtosis value greater than the second threshold and the plurality of the trips; and

rank the driver for the predetermined time window by comparing the first score and the second score of the driver with a first score and a second score of one or more drivers with an overlap in the predetermined time window between the driver and the one or more drivers; and

monitor driving behavior of the driver based upon the rank corresponding to the first score and the second score.

2. The system of claim 1, wherein the processor is further configured to execute a computer-readable instruction in order to evaluate the driving behavior of the driver with respect to one or more other drivers by comparing the first score and the second score of the driver with the first score and the second score, respectively, associated with the one or more other drivers.

3. The system of claim 1, wherein the acceleration samples are captured based upon a speed of a vehicle driven by each driver, and wherein the speed is measured using one of a vehicle speed measuring sensors comprising a GPS sensor, an accelerometer sensor, an OBD II speed measurement device or a dedicated navigation device installed in the vehicle.

4. The system of claim 1, wherein the Kurtosis values are filtered by retaining a set of Kurtosis values having corresponding Skewness values close to zero.

5. The system of claim 1, wherein the probability distribution function is determined by removing outliers from the filtered Kurtosis values and fitting the filtered Kurtosis values, after the removal of the outliers, into a normal distribution function.

6. The system of claim 1, wherein the first threshold is determined using a formula, Th=μ

+2σ

, wherein Th indicates the first threshold, μ

indicates the mean and a indicates the standard deviation.

7. The system of claim 6, wherein the first score is computed using a formula, AI=NK/NT, wherein AI indicates the first score, NK indicates the number of trips having the Kurtosis value greater than the first threshold and NT indicates the plurality of trips.

8. The system of claim 7, wherein the second score is computed using a formula, AI(N)=NN/NT, wherein AI(N) indicates the second score, NN indicates the number of trips having the Kurtosis value greater than the second threshold and NT indicates the plurality of trips.

9. A processor-implemented method for monitoring driving behavior of a driver, the method comprising:

capturing, by a processor, a plurality of acceleration samples associated with a plurality of trips completed by a plurality of drivers at a predefined time interval, wherein the plurality of acceleration samples is segregated into multiple windows such that each window comprises acceleration samples captured corresponding to a predetermined time window for each driver;

computing, by the processor, Kurtosis values and Skewness values corresponding to a set of acceleration samples, of the plurality of acceleration samples, associated with each driver;

filtering, by the processor, the Kurtosis values to obtain filtered Kurtosis values, wherein the Kurtosis values are filtered based upon the Skewness values;

determining, by the processor, a probability distribution function of the filtered Kurtosis values;

computing, by the processor, a mean and a standard deviation associated with the filtered Kurtosis values based upon the probability distribution function;

determining, by the processor, a first threshold for each driver based upon the mean and the standard deviation;

computing, by the processor, a first score for each driver based upon a number of trips having a Kurtosis value greater than the first threshold and the plurality of the trips;

determining, by the processor, a second threshold, wherein the second threshold is a minimum of the first threshold determined for each of the plurality of drivers;

computing, by the processor, a second score for each driver based upon a number of trips having a Kurtosis value greater than the second threshold and the plurality of the trips;

ranking, by the processor, the driver for the predetermined time window by comparing the first score and the second score of the driver with a first score and a second score of one or more drivers with an overlap in the predetermined time window between the driver and the one or more drivers; and

monitoring, by the processor, driving behavior of the driver based upon the rank corresponding to the first score and the second score.

10. The method of claim 9 further comprising evaluating, by the processor, the driving behavior of the driver with respect to one or more other drivers by comparing the first score and the second score of the driver with the first score and the second score, respectively, associated with the one or more other drivers.

11. A non-transitory computer readable medium embodying a program executable in a computing device for monitoring driving behavior of a driver, the program comprising:

a program code for capturing a plurality of acceleration samples associated with a plurality of trips completed by a plurality of drivers at a predefined time interval, wherein the plurality of acceleration samples is segregated into multiple windows such that each window comprises acceleration samples captured corresponding to a predetermined time window for each driver;

a program code for computing Kurtosis values and Skewness values corresponding to a set of acceleration samples, of the plurality of acceleration samples, associated with each driver;

a program code for filtering the Kurtosis values to obtain filtered Kurtosis values, wherein the Kurtosis values are filtered based upon the Skewness values;

a program code for determining a probability distribution function of the filtered Kurtosis values;

a program code for computing a mean and a standard deviation associated with the filtered Kurtosis values based upon the probability distribution function;

a program code for determining a first threshold for each driver based upon the mean and the standard deviation;

a program code for computing a first score for each driver based upon a number of trips having a Kurtosis value greater than the first threshold and the plurality of the trips;

a program code for determining a second threshold, wherein the second threshold is a minimum of the first threshold determined for each of the plurality of drivers;

a program code for computing a second score for each driver based upon a number of trips having a Kurtosis value greater than the second threshold and the plurality of the trips;

a program code for ranking the driver for the predetermined time window by comparing the first score and the second score of the driver with a first score and a second score of one or more drivers with an overlap in the predetermined time window between the driver and the one or more drivers; and

a program code for monitoring driving behavior of the driver based on the rank corresponding to the first score and the second score.