Vehicle sensor caliration for determining vehicle dynamics

• US 20110202225A1
• Filed: 02/12/2010
• Published: 08/18/2011
• Est. Priority Date: 02/12/2010
• Status: Active Grant

• Current Assignee: GeoTab Incorporated
• Original Assignee: Webtech Wireless Incorporated (GeoTab Incorporated)
• Inventors: Mclntyre, Richard Ian, McConnell, Peter Robert Henderson, West, Gerald Alexander, Willis, Richard Ernest
• Granted Patent: US 8,589,015 B2
• Time in Patent Office: Days
• US Class Current: 701/29
• CPC Class Codes: G01C 25/005 initial alignment, calibration or starting-up of inertial devices
• CPC Class Codes: G01P 21/00 Testing or calibrating of apparatus or devices covered by the preceding groups

Abstract

An accelerometer sensor equipped device uses GPS and known alignment data to determine the alignment of the accelerometer sub-system when the vehicle is stationary and in motion. The alignment data is determined from known surface information, measured GPS velocity, and measured GPS Heading.

Claims

1. A method of calibrating a tri-axial accelerometer (providing uncalibrated outputs for X, Y and Z Cartesian axes) resident in a locator that is installed on a movable asset, comprising the steps of:

a) obtaining the uncalibrated X, Y, and Z axes accelerometer output data;

b) determining a coordinate transform T that is appropriate for the tri-axial orientation of the accelerometer relative to the frame of reference of the movable asset;

c) applying said transform to said uncalibrated X, Y, and Z axes accelerometer output data to obtain calibrated X, Y, and Z axes accelerometer data.

2. The method of claim 1, wherein said transform T is of the form of:

3. The method of claim 2, wherein said transform T is obtained in auto calibration mode comprising (a) a first phase, static phase to obtain θ

(roll) and ψ

(pitch), and (b) a second, dynamic phase to obtain φ

(yaw).

4. The method of claim 2, wherein said transform T is obtained in self calibration mode comprising (a) a first phase, static phase to obtain θ

(roll) and ψ

(pitch), and (b) a second, dynamic phase to obtain φ

(yaw).

5. The method of claim 3, wherein the second dynamic phase in obtaining φ

(yaw), has the step of gating Alx and Aly acceleration samples according to one or more of the following;

a) variance of the vehicle heading is less than a preset threshold variance threshold;

b) vehicle speed greater than a preset threshold speed; and

c) magnitude of accelerations A_lx and of A_ly are greater than preset threshold values.

6. The method of claim 4, further comprising the step of stopping the process to obtain φ

(yaw) when an adjustment thereto is below a preset threshold value.

7. The method of claim 5, wherein the step of obtaining φ

(yaw) uses one of the following algorithms {Gradient Descent, Least Mean Squares, Recursive Least Squares, Kalman, Fast Kalman}.

8. The method of claim 3, further comprising the step of compensating for slope that is performed between said static phase and said dynamic phase.

9. The method of claim 3 for filtering and thereby qualifying events of interest, comprising the steps of:

(a) configuring an envelope of acceleration data to correspond to an event of interest;

(b) evaluating samples of acceleration data and determining if said samples fit over said envelope; and

(c) inferring an event of interest based on said evaluating.

10. The method of claim 3, comprising the use of GPS, GLONAS, or other positioning systems to obtain time, location, velocity, altitude, and heading of said mobile asset.

11. The method of claim 3, comprising the use of wireless modem interfaces GSM, GPRS, CDMA, HSDPA, HSPA+, HSUPA, UMTS, EDGE, Iridium, Wi-Fi, Bluetooth, Zigbee, or proprietary wireless data networks.

12. The method of claim 3, wherein the tri-axial accelerometer is implemented by the combination of a two axis accelerometer to obtain θ

and ψ

(pitch and roll) and a single axis accelerometer to obtain φ

(yaw).

13. The method of claim 3, wherein the tri-axial accelerometer is implemented by the combination of a single axis accelerometer to obtain θ

(roll), a single axis accelerometer to obtain ψ

(pitch) and a single axis accelerometer to obtain φ

(yaw).

14. The method of claim 3, wherein the vehicle is equipped with a gyroscope, and the further step of fusing acceleration data with data from said gyroscope.

15. The method of claim 3, further comprising the step of developing metrics for aggressive driving and the step of using the calibrated acceleration data to determine if the vehicle exceeds said metrics.

16. The method of claim 3, further comprising means for removing from acceleration data, the offset created by gravity when the mobile asset is moving on a sloped terrain, comprising the steps of:

(a) low pass filter the z-axis acceleration data to obtain â

z;

(b) calculate the terrain pitch;

ρ

=arc cos(â

z/G);

(c) determine the sign (negative or positive) of the slope; and

(d) correct said filtered the x-axis acceleration data by subtracting therefrom G·

sin(ρ

).

17. The method of claim 3, further comprising means for removing from acceleration data, the offset created by gravity when the mobile asset is moving on a sloped terrain, comprising the step of:

s

(a) high pass filter the acceleration data a_x.

18. A method of monitoring aggressive driving operation of a mobile asset, comprising the steps of:

(a) obtaining calibrated acceleration data A_vx and A_vy; and

(b) determining the range chosen among;

Normal Range;

|A_vx|<

0.39 G and |A_vy|<

0.45 G

Cautionary Range;

0.39 G<

|A_vx|<

0.5 G or 0.45 G<

|A_vy|<

0.7 G

Aggressive Range;

|A_vx|>

0.50 G or |A_vy|>

0.70 G

where said obtained calibrated acceleration data lies within.

20. The method of claim 18, further comprising the step of storing calibrated acceleration data to establish a temporal record of the acceleration dynamics and the aggressive driving metrics for use by insurers.

19. A method of monitoring aggressive driving operation of a mobile asset, comprising the steps of combining speed and acceleration data when speed data is only derivable from acceleration data, comprising the steps of:

(a) the asset experiences a longitudinal deceleration of greater than 0.5 G over a period of 2.2 seconds, and

(b) the assert velocity decreased over a period of 2.2 seconds from a minimum of 35 km/h to near zero.