Motion classification methods for personal navigation

• US 7,561,960 B2
• Filed: 09/22/2006
• Issued: 07/14/2009
• Est. Priority Date: 04/20/2006
• Status: Expired due to Fees

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Soehren, Wayne A.
• Primary Examiner(s): ARTHUR JEANGLAUDE, GERTRUDE
• Application Number:
  US11/534,511
• Publication Number:
  US 20070250261A1
• Time in Patent Office: 1,026 Days
• Field of Search: 701/200, 701/207, 701208-211, 340/988, 340/995.1, 600/595, 702/160
• US Class Current: 701/207
• CPC Class Codes: G01C 21/188 for accumulated errors, e.g. by coupling inertial systems with absolute positioning systems
• CPC Class Codes: G01C 22/006 Pedometers
• CPC Class Codes: G01C 25/005 initial alignment, calibration or starting-up of inertial devices

Abstract

A personal navigation system including one or more sensors that sense motion of a human and output signals corresponding to the motion of the human and a human-motion-classification processing block that receives sensor data from the one or more sensors. The human-motion-classification processing block includes a Kalman filter processing block, an inertial navigation processing block, and a motion classification processing block. The Kalman filter processing block executes a Kalman filter that provides corrections to the motion classification processing block. The inertial navigation processing block receives input sensor data from the sensors and outputs a navigation solution. The motion classification processing block executes a motion classification algorithm that implements a step-time threshold between two types of motion, identifies a type of motion based on the received sensor data, and outputs a distance-traveled estimate to the Kalman filter processing block based on the identified type of motion.

Claims

1. A personal navigation system, the system comprising:

one or more sensors adapted to sense motion of a human and to output signals corresponding to the motion of the human;

a human-motion-classification processing block adapted to receive sensor data from the one or more sensors, the human-motion-classification processing block comprising;

a Kalman filter processing block adapted to execute at least one Kalman filter;

an inertial navigation processing block adapted to receive input sensor data from the one or more sensors and to output a navigation solution; and

a motion classification processing block adapted to execute a motion classification algorithm that implements a step-time threshold between two types of motion, the step-time threshold determined from running model parameters, walking model parameters, and a threshold speed, wherein the motion classification processing block is further adapted to identify a type of motion, and to output a distance-traveled estimate to the Kalman filter processing block based on the identified type of motion, wherein the at least one Kalman filter provides corrections to the motion classification processing block.

2. The system of claim 1, the human-motion-classification processing block further comprising:

a terrain correlation processing block adapted to execute a terrain correlation algorithm to provide an estimate of terrain correlated to a position of the human in order to bound the navigational error of the personal navigation system.

3. The system of claim 1, wherein the types of motion comprise walking forward, walking backward, running forward, stopped, and walking sideways.

4. The system of claim 1, wherein the sensors comprise:

one or more global positioning systems, one or more accelerometers, one or more gyroscopes, one or more magnetic sensors, one or more pressure sensors, one or more altimeters, one or more inertial sensors, and combinations thereof.

5. The system of claim 1, wherein the human-motion -classification processing block further comprises:

a calibration processing block to calibrate the step-time threshold, wherein the running model parameters, the walking model parameters, and the threshold speed are calibrated for each user.

6. A method for estimating foot-travel position, the method comprising:

receiving sensor data at a motion classification algorithm having calibrated computed models for types of motion calibrated to a user;

determining a step length estimate at the motion classification algorithm; and

receiving corrections to the step length model from a Kalman filter, wherein the calibrated computed models for types of motion include running model parameters, and walking model parameters that are used to determine a step-time threshold.

7. The method of claim 6, wherein receiving corrections to the distance -traveled model from a Kalman filter comprises:

detecting erroneous results in the step length estimate at the Kalman filter;

rejecting the step length estimate based on the detected erroneous results; and

operating a personal navigation system without corrections based on the rejecting.

8. The method of claim 6, further comprising:

calibrating initial models for types of motion for the user; and

providing the calibrated models in the motion classification algorithm.

9. The method of claim 8, wherein calibrating models for types of motion for the user comprises:

receiving a first type-of motion input to initialize a calibrating process for the first type of motion;

providing sensor data indicative of the first type of motion by the user during the calibrating process for the first type of motion;

computing models for the first type of motion;

receiving a second type-of motion input to initialize a calibrating process for the second type of motion;

providing sensor data indicative of the second type of motion by the user during the calibrating process for the second type of motion; and

computing models for the second type of motion.

10. The method of claim 9, wherein receiving a first type-of motion input comprises receiving a walking input at a calibration processing block, and wherein computing models for the first type of motion comprises:

computing a time period of a walking step indicative of a time elapsed between foot falls while the user walks, wherein the computing a time period of a walking step is based on sensor data received at the calibration processing block;

estimating a walking step length based on the time period of a walking step and a speed of walking at the calibration processing block; and

determining walking model parameters based on the walking step length and the time period of a walking step at the calibration processing block.

11. The method of claim 10, wherein receiving a second type-of-motion input comprises receiving a running input at a calibration processing block, and wherein computing models for the second type of motion comprises:

computing a time period of a running step indicative of a time elapsed between foot falls while the user runs, wherein the computing a time period of a running step is based on sensor data received at the calibration processing block;

estimating a running step length based on the time period of a running step and a speed of running; and

determining running model parameters based on the running step length and the time period of a running step.

12. The method of claim 11, the method further comprising:

determining the step-time threshold between walking and running based on the providing calibrated computed models.

13. The method of claim 6, the method further comprising:

determining a navigation heading of the user;

determining a direction of motion of the user; and

determining if an angular difference between the navigation heading and the direction of motion is within a selected range of angles

14. The method of claim 13, the method further comprising:

determining the user is stepping backward if the navigation heading and the direction of motion are substantially opposite each other and within a selected range of opposing angles.

15. A processor readable medium having programmable-processor executable instructions for performing a method comprising:

receiving sensor data at a motion classification algorithm having calibrated computed models for types of motion calibrated to a user;

determining a step length estimate at the motion classification algorithm; and

receiving corrections to the step length model from a Kalman filter, wherein the calibrated computed models for types of motion include running model parameters, and walking model parameters that are used to determine a step-time threshold.

16. The medium of claim 15, wherein the programmable-processor executable instructions for receiving corrections to the distance-traveled estimate from a Kalman filter comprise programmable-processor executable instructions for:

detecting erroneous results in the step length estimate at the Kalman filter;

rejecting the step length estimate based on the detected erroneous results;

operating a personal navigation system without corrections based on the rejecting;

calibrating models for types of motion for the user; and

providing the calibrated models in the motion classification algorithm.

17. The medium of claim 16, wherein the programmable-processor executable instructions for calibrating models for types of motion for the user comprise programmable-processor executable instructions for:

receiving a first type-of-motion input to initialize a calibrating process for the first type of motion;

providing sensor data indicative of the first type of motion by the user during the calibrating process for the first type of motion;

computing models for the first type of motion;

receiving a second type-of-motion input to initialize a calibrating process for the second type of motion;

providing sensor data indicative of the second type of motion by the user during the calibrating process for the second type of motion; and

computing models for the second type of motion.

18. The medium of claim 17, wherein the programmable-processor executable instructions for receiving a first type-of-motion input comprise programmable-processor executable instructions for:

receiving a walking input at a calibration processing block, and wherein the programmable-processor executable instructions for computing models for the first type of motion comprise programmable-processor executable instructions for;

computing a time period of a walking step indicative of a time elapsed between foot falls while the user walks, wherein the computing a time period of a walking step is based on sensor data received at the calibration processing block;

estimating a walking step length based on the time period of a walking step and a speed of walking at the calibration processing block; and

determining walking model parameters based on the walking step length and the time period of a walking step at the calibration processing block.

19. The medium of claim 18, wherein the programmable-processor executable instructions for receiving a second type-of-motion input comprise programmable-processor executable instructions for:

receiving a running input at a calibration processing block and wherein, the programmable-processor executable instructions for computing models for the second type of motion comprise programmable-processor executable instructions for;

computing a time period of a running step indicative of a time elapsed between foot falls while the user runs, wherein the computing a time period of a running step is based on sensor data received at the calibration processing block;

estimating a running step length based on the time period of a running step and a speed of running;

determining running model parameters based on the running step length and the time period of a running step, and wherein the medium further comprises programmable-processor executable instructions for determining the step-time threshold between walking and running based on the providing calibrated computed models.

20. The medium of claim 15, further comprising programmable-processor executable instructions for:

determining a navigation heading of the user;

determining a direction of motion of the user;

determining if an angular difference between the navigation heading and the direction of motion is within a selected range of angles; and

determining the user is stepping backward if the navigation heading and the direction of motion are substantially opposite each other and within a selected range of opposing angles.