# US 10,502,576 B2 · Determining the position of a mobile device in a geographical area

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 10,502,576 B2
- Document ID: US10502576B2
- Status: Granted
- Kind: B2
- Application number: 15/509,771
- Filed: 2015-09-10
- Published: 2017-08-24
- Granted: 2019-12-10
- Earliest claimed: 2014-09-15
- Assignees: Oxford University Innovation Limited
- Applicants: Oxford University Innovation Limited
- Inventors: Agathoniki Trigoni; Zhuoling Xiao
- Web page: https://patentgrep.com/patent/US10502576B2

## CPC

- G01C 21/206: {specially adapted for indoor navigation}
- G01C 21/12: executed aboard the object being navigated; Dead reckoning
- G01C 22/006: {Pedometers}
- G01C: MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY (measuring liquid level G01F; radio navigation, determining distance or velocity by use of propagation effects, e.g. Doppler effects, propagation time, of radio waves, analogous arrangements using other waves G01S)

## Abstract

A mobile device (100) carried by a user has sensors providing acceleration and orientation information. These are processed to provide trajectory for the device. The processing includes improved methods using different modules (1, 2, 3,5, 6, 7) for classifying the motion (1) and detecting a user's steps (5), tracking the orientation (3) and acceleration (2) of the device (100), estimating the length of the user's steps (6) and estimating the heading (7) of the user. The trajectories are compared to a map and corrected by the constraints provided by the map. Parameters of the methods of producing the trajectories are optimised on the basis of maximising probability of the trajectory found by the map matching given the indicated trajectory.

## Claims

### Claim 1

A method of optimising an orientation error for a device carried by a user, the orientation error being the difference between an estimated orientation calculated from signals from one or more sensors of the device and the real orientation, the method comprising:
receiving acceleration and orientation signals from the one or more sensors responsive to the acceleration and orientation of the device,
classifying motion of the device as a symmetric motion mode or an asymmetrical motion mode, the symmetric motion mode having similar periodic acceleration signals with a period corresponding to a single step of the user, and the asymmetrical motion mode having similar periodic orientation patterns with a period corresponding to two steps of the user,
obtaining a set of vector values of the acceleration of the device from the acceleration signals for an interval of one or more whole periods of a single step, or a two steps, of the user, according to the classification of the motion of the device as the symmetrical motion mode or the asymmetrical motion mode, and
optimising the orientation error using the set of acceleration values subject to physical constraints on the motion that apply during the interval.

### Claim 2 (depends on 1)

The method as claimed in claim 1 wherein the physical constraints include that the orientation error is constant during the interval.

### Claim 3 (depends on 1)

The method as claimed in claim 1 wherein the physical constraints include that the device is the same at the beginning and end of the interval.

### Claim 4 (depends on 1)

The method as claimed in claim 1, wherein the physical constraints include that the vertical velocity, with respect to the Earth, of the device does not change over the interval.

### Claim 5 (depends on 1)

The method as claimed in claim 1, wherein the physical constraints include that the horizontal velocity, with respect to the Earth, of the device does not change over the interval.

### Claim 6 (depends on 1)

The method as claimed in claim 1, wherein the physical constraints include that the 3D velocity, with respect to the Earth, of the device does not change over the interval.

### Claim 7 (depends on 1)

The method as claimed in claim 1 wherein the optimising is by regression.

### Claim 8 (depends on 7)

The method as claimed in claim 7 wherein the optimising is by the method of least squares.

### Claim 9 (depends on 7)

The method as claimed in claim 7 wherein the optimising is Bayesian regression.

### Claim 10 (depends on 1)

The method as claimed in claim 1 wherein the optimising comprises optimising the following equations for R:
Δv=ve−v0=R·Σt=t0teat·Δt=0 and
Δl=R·Σt=t0te(Στ=t0tatz·Δt)·Δt=0,
where at are the vector acceleration values for the device, t is time, times t0 and te are respective times at the beginning and end of the interval, v0 and ve are velocity values at times t0 and te respectively, Δt and Δτ are the interval between acceleration values, az is the acceleration along the vertical axis, Δl is the change in height of the device over the interval, Δv is the change in velocity over the interval, and R is a matrix representing the orientation error.

### Claim 11 (depends on 1)

The method as claimed in claim 1 wherein the number of whole periods of a step, or a double step, of the interval is one.

### Claim 12 (depends on 1)

The method as claimed in claim 1 wherein the coordinates in which the acceleration is expressed are those of a coordinate system fixed with respect to the Earth.

### Claim 13 (depends on 1)

A method of correcting acceleration signals obtained from sensors of a device, the method comprising:
performing the method of optimising an orientation error according to claim 1, and
applying the orientation error to acceleration signals of the interval.

### Claim 14

A method of determining the position of a mobile device in a geographical area, the device having one or more sensors that provide information about the motion of the device, wherein the position is determined using detected steps of a user carrying the device, and wherein the method comprises the steps of:
detecting steps of the user carrying the device from the signals from the sensors by detecting a pattern of boundaries between steps using the signals from the sensors,
classifying a detected step as a real step of the user, or a fake step where the user does not take a real step,
wherein the classification is performed by: comparing features of signals from the sensors indicative of one or more of acceleration, velocity, displacement, and orientation at times related to that step.

### Claim 15 (depends on 14)

The method as claimed in claim 14 wherein the comparing includes a feature of the horizontal displacement signal.

### Claim 16 (depends on 14)

The method as claimed in claim 14 wherein the comparing includes a feature of the vertical displacement signal.

### Claim 17 (depends on 14)

The method as claimed in claim 14 wherein the comparing includes comparing the horizontal displacement signal to the vertical displacement signal.

### Claim 18 (depends on 17)

The method as claimed in claim 17 wherein the comparing includes deciding which of the horizontal and vertical displacements is larger.

### Claim 19 (depends on 17)

The method as claimed in claim 17 wherein at least one of the vertical and/or horizontal displacement signals are formed by double integrating the acceleration signals.

### Claim 20 (depends on 14)

The method as claimed in claim 14 wherein the comparing includes cross correlating the orientation with the acceleration signals.

### Claim 21 (depends on 14)

The method as claimed in claim 14 wherein the comparing includes cross correlating the orientation with the velocity or displacement signals.

### Claim 22 (depends on 21)

The method as claimed in claim 21 wherein the velocity or displacement signals are formed by respectively single or double integration of the acceleration signal.

### Claim 23 (depends on 14)

The method as claimed in claim 14 wherein the comparing includes examining an orientation offset, that being the change in orientation between the beginning and end of a step of the user.