MOTION CLASSIFICATION METHODS FOR PERSONAL NAVIGATION
First Claim
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, wherein the motion classification processing block is further adapted to identify a type of motion based on the received sensor data, 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.
1 Assignment
0 Petitions
Accused Products
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.
146 Citations
20 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, wherein the motion classification processing block is further adapted to identify a type of motion based on the received sensor data, 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. - View Dependent Claims (2, 3, 4, 5)
-
-
6. A method for estimating foot-travel position, the method comprising:
-
receiving sensor data at a motion classification algorithm having a motion type threshold 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. - View Dependent Claims (7, 8, 9, 10, 11, 12, 13, 14)
-
-
15. A programmable processor readable medium readable medium having programmable-processor executable instructions for performing a method comprising:
-
receiving sensor data at a motion classification algorithm having a motion type threshold 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. - View Dependent Claims (16, 17, 18, 19, 20)
-
Specification