Optical inertial measurement apparatus and method

Patent metadata

Field	Value
Publication number	US-2017307380-A1
Application number	US-201515520580-A
Country	US
Kind code	A1
Filing date	Jan 30, 2015
Priority date	Oct 20, 2014
Publication date	Oct 26, 2017
Grant date	—

Abstract

An optical inertial measurement method for determining a 6DOF pose in respect of a moving platform, the method comprising providing, in respect of said moving platform, a camera unit ( 10 ) comprised of at least three monocular cameras ( 14, 16, 18 ) connected rigidly together and configured such that their fields of view do not overlap and cover motion of said platform in each of three principal, substantially orthogonal axes; receiving video outputs from each of said cameras; determining individual point correspondences from said video outputs, and estimating therefrom, for each camera, a transform that includes translation values representative of direction of translation in the x and y axes, rotation about the optical axis and a scale factor, each transform being expressed with respect to a local 3D coordinate system associated with a respective camera; and mapping said translation values in the x and y axes to a global 3D coordinate system, having its origin defined by a point between said cameras, and multiplying said translation values by a respective scale factor so as to determine a 6DOF pose in respect of said camera unit.
An optical inertial measurement system is also disclosed.

First claim

1 . An optical inertial measurement method for determining a 6DOF pose in respect of a moving platform, the method comprising:
receiving video outputs from each of at least three monocular cameras, said cameras included in a camera unit and spatially fixed with respect to one another, such that their fields of view do not overlap and cover motion of said platform along each of three principal, substantially orthogonal axes; determining individual point correspondences from said video outputs, and estimating therefrom, for each camera, a transform that includes translation values representative of direction of translation in x and y axes of a respective camera, rotation about an optical axis of a respective camera, and a scale factor of a respective camera, each transform being expressed with respect to a local 3D coordinate system associated with a respective camera; and mapping said translation values in the x and y axes to a global 3D coordinate system, having its origin defined by a point between said cameras, and multiplying said translation values by a respective said scale factor so as to determine a 6DOF pose in respect of said camera unit.

2 . The method according to claim 1 , wherein said camera unit comprises three monocular cameras rigidly connected together, wherein the optical axis of each of said three cameras is aligned with a respective one of said three principal orthogonal axes.

3 . The method according to claim 1 , wherein a first of said cameras has its optical axis aligned with a first one of said three principal orthogonal axes, and a plurality of further monocular cameras are configured to capture non-overlapping views from along the remaining two principal axes.

4 . The method according to claim 1 , wherein the local 3D coordinate system for each camera is a Cartesian coordinate system having its origin at an optical centre of the respective camera.

5 . The method according to claim 4 , wherein the z-axis of said Cartesian coordinate system coincides with the optical axis of the respective camera.

6 . The method according to claim 1 , wherein said global 3D coordinate system has its origin at a centre point between said cameras.

7 . An optical inertial measurement apparatus for determining a 6DOF pose in respect of a moving platform, the apparatus comprising:
a camera unit comprised of at least three monocular cameras connected rigidly together and configured such that their fields of view do not overlap and cover motion of said platform along each of three principal, substantially orthogonal axes; and a central processing unit for receiving video outputs from each of said cameras, and being configured to
determine individual point correspondences from said video outputs and estimate therefrom, for each camera, a transform that includes translation values representative of direction of translation in x and y axes of a respective camera, rotation about an optical axis of a respective camera, and a scale factor of a respective camera, each transform being expressed with respect to a local 3D coordinate system associated with a respective camera, and
map said translation values in the x and y axes to a global 3D coordinate system, having its origin defined by a point between said cameras, and multiply said translation values by a respective said scale factor so as to determine a 6DOF pose in respect of said camera unit.

8 . The apparatus according to claim 7 , wherein said camera unit comprises three monocular cameras spatially fixed with respect to one another, wherein the optical axis of each of said three cameras is aligned with a respective one of said three principal orthogonal axes.

9 . The apparatus according to claim 7 , wherein a first of said cameras has its optical axis aligned with a first one of said three principal orthogonal axes, and a plurality of further monocular cameras are configured to capture non-overlapping views from along the remaining two principal axes.

10 . The apparatus according to claim 7 , wherein the local 3D coordinate system for each camera is a Cartesian coordinate system having its origin at an optical centre of the respective camera.

11 . The apparatus according to claim 10 , wherein the z-axis of said Cartesian coordinate system coincides with the optical axis of the respective camera.

12 . The apparatus according to claim 7 , wherein said global 3D coordinate system has its origin at a centre point between said cameras.

13 . An optical inertial measurement apparatus, comprising:
a camera unit comprised of three monocular cameras spatially fixed with respect to one another, such that their fields of view do not overlap and cover motion of said platform along each of three principal, substantially orthogonal axes; and a central processing unit for receiving video outputs from each of said cameras, and being configured to
determine individual point correspondences from said video outputs and estimate therefrom, for each camera, a transform that includes translation values representative of direction of translation in x and y axes of a respective camera, rotation about an optical axis of a respective camera, and a scale factor of a respective camera, each transform being expressed with respect to a local 3D coordinate system associated with a respective camera, wherein the optical axis of each of said three cameras is aligned with a respective one of said three principal orthogonal axes, and
map said translation values in the x and y axes to a global 3D coordinate system, having its origin defined by a point between said cameras, and multiply said translation values by a respective said scale factor so as to determine a 6DOF pose in respect of said camera unit.

14 . The apparatus according to claim 13 , wherein a first of said cameras has its optical axis aligned with a first one of said three principal orthogonal axes, and a plurality of further monocular cameras are configured to capture non-overlapping views from along the remaining two principal axes.

15 . The apparatus according to claim 14 , wherein the local 3D coordinate system for each camera is a Cartesian coordinate system having its origin at an optical centre of the respective camera.

16 . The apparatus according to claim 13 , wherein the local 3D coordinate system for each camera is a Cartesian coordinate system having its origin at an optical centre of the respective camera.

17 . The apparatus according to claim 16 , wherein said global 3D coordinate system has its origin at a centre point between said cameras.

18 . The apparatus according to claim 16 , wherein the z-axis of said Cartesian coordinate system coincides with the optical axis of the respective camera.

19 . The apparatus according to claim 18 , wherein said global 3D coordinate system has its origin at a centre point between said cameras.

20 . The apparatus according to claim 13 , wherein said global 3D coordinate system has its origin at a centre point between said cameras.

Assignees

• Bae Systems Plc

Inventors

• Revell James Duncan
• Goodall Mark Robert
• Cross Gary Martin

Classifications

• G01C21/16 Primary

  by integrating acceleration or speed, i.e. inertial navigation · CPC title

• G06T2207/30244

  Camera pose · CPC title

• G06T7/70

  Determining position or orientation of objects or cameras (camera calibration G06T7/80) · CPC title

• G01C21/1656 Primary

  with passive imaging devices, e.g. cameras · CPC title

• G01C21/165 Primary

  combined with non-inertial navigation instruments · CPC title