System and method for vehicle driver behavior analysis and evaluation

• US 7,389,178 B2
• Filed: 07/20/2004
• Issued: 06/17/2008
• Est. Priority Date: 12/11/2003
• Status: Active Grant

• Current Assignee: Greenroad Driving Technologies Limited
• Original Assignee: Greenroad Driving Technologies Limited
• Inventors: Mulchadsky, Itamar, Fleishman, Hod, Raz, Ofer
• Primary Examiner(s): ARTHUR JEANGLAUDE, GERTRUDE
• Application Number:
  US10/894,345
• Publication Number:
  US 20050131597A1
• Time in Patent Office: 1,428 Days
• Field of Search: 701/1, 701/29, 701/31, 701 33- 34, 701/36, 701/56, 701/301, 340/438, 340/439, 340/903
• US Class Current: 701/200
• CPC Class Codes: G07C 5/085 using electronic data carriers
• CPC Class Codes: G09B 19/167 Control of land vehicles

Abstract

A system and method for analyzing and evaluating the performance and attitude of a motor vehicle driver. A raw data stream from a set of vehicle sensors is filtered to eliminate extraneous noise, and then parsed to convert the stream into a string of driving event primitives. The string of driving events is then processed by a pattern-recognition system to derive a sequence of higher-level driving maneuvers. Driving maneuvers include such familiar procedures as lane changing, passing, and turn and brake. Driving events and maneuvers are quantified by parameters developed from the sensor data. The parameters and timing of the maneuvers can be analyzed to determine skill and attitude factors for evaluating the driver'"'"'s abilities and safety ratings. The rendering of the data into common driving-related concepts allows more accurate and meaningful analysis and evaluation than is possible with ordinary statistical threshold-based analysis.

Claims

1. A system for analyzing and evaluating the performance and behavior of a driver of a vehicle, the system comprising:

(a) a vehicle sensor utility operative to monitor the state of the vehicle and to output a raw data stream corresponding thereto;

(b) a driving event handler operative to receive the raw data stream, detect from said raw data stream a plurality of distinctive patterns each associated with a corresponding driving event representation and to output a driving event string containing said driving event representations and their corresponding values; and

(c) a maneuver detector operative to receive said driving event string, compare a sequence of said driving event representations comprised in the driving event string with patterns of driving maneuvers comprised in a database, to enable recognizing patterns of driving maneuvers in said driving event string and to construct and output a driving maneuver representation corresponding thereto, said driving maneuver representation containing a representation of at least one driving maneuver.

2. The system of claim 1, further comprising:

(d) a plurality of vehicle sensors operative to monitor different states of the vehicle, wherein each of said vehicle sensors is operative to output a raw data stream.

3. The system of claim 1, wherein each of the plurality of distinctive patterns is selected from among a group consisting of:

(i) a start event;

(ii) an end event;

(iii) a maximum event;

(iv) a minimum event;

(v) a cross event;

(vi) a flat event;

(vii) a local maximum event; and

(viii) a local flat event.

4. The system of claim 1, further comprising:

(d) a skill assessor operative to analyzing the skill of the driver based upon said at least one driving maneuver.

5. The system of claim 1, further comprising:

(d) an attitude assessor operative to analyzing the attitude of the driver based upon said at least one driving maneuver.

6. The system of claim 1, further comprising:

(d) a database operative to record characteristic driving maneuvers of the driver; and

(e) an anomaly detector operative to compare said characteristic driving maneuvers to said at least one driving maneuver.

7. The system of claim 1, wherein said driving event string provided to said maneuver detector is a compressed representation of said raw data stream.

8. The system of claim 1, wherein said driving event string is transmitted by wireless from the vehicle to said maneuver detector.

9. The system of claim 1, further comprising evaluating means operative to receive values corresponding to a sequence of driving events representations associated with a driving maneuver and derive therefrom an overall risk rating that reflects the risk experienced while carrying out said driving maneuver, and evaluate the performance of the driver of the vehicle based on said risk rating.

10. A method for analyzing and evaluating the performance and behavior of a driver of a vehicle, based on a raw data stream from a set of sensors operative to monitor the state of the vehicle, the method comprising:

(a) detecting from said raw data stream a plurality of distinctive patterns each associated with a corresponding driving event representation and generating a driving event string containing said driving event representations and their corresponding values; and

(b) comparing a sequence of said driving events representations comprised in the driving event string with patterns of driving maneuvers comprised in a database, recognizing patterns of driving maneuvers in said driving event string and generating a driving maneuver representation containing a representation of at least one driving maneuver.

11. The method of claim 10, furthermore comprising:

(c) assessing driver skill, based on said driving maneuvers.

12. The method of claim 10, furthermore comprising:

(c) assessing driver attitude, based on said driving maneuvers.

13. The method of claim 10, wherein each of the plurality of distinctive patterns is selected from among a group consisting of:

start event, an end event, a maximum event, a minimum event, a cross event, a local maximum event, a local minimum event, a local flat event and a flat event.

14. The method of claim 10, wherein said driving event string is a compressed representation of said raw data stream.

15. The method of claim 10, further comprising a step of receiving values corresponding to a sequence of driving events representations associated with a driving maneuver and deriving therefrom an overall risk rating that reflects the risk experienced while carrying out said driving maneuver, and evaluating the performance of the driver of the vehicle based on said risk rating.