Search

EP-3864375-B1 - A METHOD OF ESTIMATING A METRIC OF INTEREST RELATED TO THE MOTION OF A BODY

EP3864375B1EP 3864375 B1EP3864375 B1EP 3864375B1EP-3864375-B1

Inventors

  • FARAGHER, RAMSEY MICHAEL
  • CROCKETT, ROBERT MARK
  • DUFFETT-SMITH, PETER

Dates

Publication Date
20260506
Application Date
20191011

Claims (15)

  1. A computer-implemented method performed in a tracking system for tracking the motion of a body, as a function of time, the method comprising: (a) during a first time period, obtaining first data related to the motion of a body from at least one primary positioning unit (20), wherein said at least one primary positioning unit is mounted on a first platform (100) carried on the body, said primary positioning unit being operational during the first time period, and wherein the at least one primary positioning unit is configured to provide position and navigation data directly; (b) during the first time period, obtaining second data from one or more secondary sensors (12', 14', 16', 18') configured to make measurements which do not provide positioning, tracking or navigation data directly, but from which position or movement may be determined, whereby said second data comprises measurements from which position or movement may be determined, said one or more secondary sensors being mounted on one or more second platforms (110) carried on the body, wherein the first and second platforms are moveable independently of each other; (c) generating (203) first training data comprising the first data and second data, and wherein the training data comprises at least one training metric related to the motion of the body during the first time period, determined using at least the first data obtained from the at least one primary positioning unit; (d) training (204) a first algorithm (46) using the first training data; (e) during a second time period, obtaining (301) third data from the one or more secondary sensors, said third data comprising measurements which do not provide positioning, tracking or navigation data directly, but from which position or movement may be determined; (f) analysing (304) the third data to estimate at least one first metric related to the motion of the body during the second time period using the first algorithm (46) ; and (g) further analysing (305) the third data to estimate the motion of the body, said estimation being constrained by the estimated at least one first metric.
  2. The method of claim 1, wherein the first data and second data are obtained for a plurality of motion contexts for the body, and wherein a motion context of the body during the second time period corresponds substantially to a motion context during the first time period.
  3. The method of claim 1 or claim 2, wherein the first data and second data are obtained for a plurality of position contexts for the one or more second platforms with respect to the body, and wherein a position context for the one or more second platforms with respect to the body during the second time period corresponds substantially to a position context during the first time period.
  4. The method of any of the preceding claims, wherein the estimated at least one first metric in step (f) is at least one of the following: a direction of motion, a speed, a velocity, a motion context for the body, and a position context for the one or more second platforms with respect to the body.
  5. The method of any of the preceding claims, wherein the first algorithm comprises a neural network.
  6. The method of any of the preceding claims, further comprising analysing the third data to estimate the evolution of at least one second metric related to the motion of the body during the second time period, wherein the evolution of the at least one second metric is constrained by the at least one first metric estimated using the first algorithm, preferably wherein the third data are analysed in order to estimate a trajectory of the body during the second time period.
  7. The method of claim 6, wherein the analysing the third data comprises comparing said third data with the at least one first metric to obtain corrected third data, and wherein the estimation of the evolution of the second metric is based on said corrected third data, preferably wherein the obtaining corrected third data comprises determining a measurement bias of the one or more secondary sensors and correcting for said measurement bias in order to obtain the corrected third data.
  8. The method of any of the preceding claims, wherein; in step (f) the third data are arranged as a plurality of frames, each frame comprising a time-ordered plurality of measurement values from the one or more secondary sensors, and wherein the first algorithm is used to provide an estimate of the at least one first metric for each of said frames, preferably wherein in step (c) the second data are arranged as a plurality of frames, each frame comprising a time-ordered plurality of measurement values from the at least one secondary sensor, and wherein; the temporal length of the frames of second data and frames of third data are substantially the same.
  9. The method of claim 8, wherein the each of the plurality of frames has the same temporal length.
  10. The method claim 8, wherein the plurality of frames have a temporal length based on at least one of a determined position context and a determined motion context during the respective first and second time periods.
  11. The method of any of the preceding claims, wherein the first algorithm is selected from a set of predetermined algorithms for estimating the at least one first metric, wherein the selection is based on the at least one first metric to be determined.
  12. The method of any of the preceding claims, further comprising the step of determining the evolution of the training metric related to the motion of the body during the first time period, and wherein the first training data comprises said evolution of the training metric, preferably wherein the first training data comprises a trajectory of the body during the first time period, preferably wherein the evolution of the training metric is determined based on the first data and second data obtained during the first time period.
  13. The method of any of the preceding claims, further comprising the steps of: (h) during a third time period, obtaining fifth data related to the motion of the body from the at least one primary positioning unit; (i) during the third time period, obtaining sixth data from the one or more secondary sensors; (j) determining the evolution of a second training metric related to the motion of the body during the third time period, based at least in part on the fifth data; (k) analysing the sixth data to estimate the evolution of the second training metric during the third time period, wherein the estimated evolution of the second training metric is constrained by at least one first metric estimated using the first algorithm trained using the first training data; (l) comparing the determined and estimated evolutions of the second training metric, and; if the difference between the determined and estimated evolutions of the second training metric is greater than a predetermined threshold, updating the first training data with second training data, wherein the second training data comprises the fifth data and sixth data.
  14. A tracking system for tracking the motion of a body, as a function of time, the system comprising: at least one primary positioning unit (20) mounted on a first platform (100) able to be carried on a body, and wherein the at least one primary positioning unit is configured to provide position and navigation data directly; one or more secondary sensors (12', 14', 16', 18') configured to make measurements which do not provide positioning, tracking or navigation data directly, but from which position or movement may be determined, mounted on one or more second platforms (110) able to be carried on a body, and; a processor adapted to perform the steps of: (a) during a first time period when the at least one primary positioning unit is operational, obtaining first data from said at least one primary positioning unit, wherein during the first time period the at least one primary positioning unit is mounted on the first platform carried on the body; (b) during the first time period, obtaining second data from the one or more secondary sensors, said second data comprising measurements from which position or movement may be determined, said one or more second platforms being carried on the body during the first time period, wherein the first and second platforms are moveable independently of each other; (c) generating (203) first training data comprising the first and second data, and wherein the training data comprises at least one training metric related to the motion of the body during the first time period, determined using at least the first data obtained from the at least one primary positioning unit; (d) training (204) a first algorithm (46) using the first training data; (e) during a second time period during which the one or more second platforms are carried on the body, obtaining (301) third data from the one or more secondary sensor, said third data comprising measurements which do not provide positioning, tracking or navigation data directly, but from which position or movement may be determined; (f) analysing (304) the third data to estimate at least one first metric related to the motion of the body during the second time period using the first algorithm (46); and (g) further analysing (305) the third data to estimate the motion of the body, said estimation being constrained by the estimated at least one first metric.
  15. The method or system of any of the preceding claims, wherein the one or more secondary sensors comprises at least one of: an accelerometer, a gyroscope, a magnetometer, a barometer, a pedometer, a light sensor, a pressure sensor, a strain sensor, a proximity sensor and a camera, and/or wherein the at least one primary positioning unit comprises at least one of: a GNSS unit, camera, a RADAR and a LIDAR.

Description

FIELD OF THE INVENTION The present invention is directed to a method and system for estimating a metric of interest related to the motion of a body. The invention has particular application in the field of tracking and navigation. BACKGROUND TO THE INVENTION A traditional inertial navigation system uses standard mechanics equations to convert measurements from inertial sensors (for example accelerometers, gyroscopes etc.) into tracking and navigation data, such as velocity, orientation and position. A known problem with such a system is that of numerical integration error, which manifests as rapidly increasing errors in the derived tracking and navigation data. The inertial navigation equations typically involve the determination of orientation, velocity and position through numerical integration of successive rotation-rate and acceleration measurements. The integration of measurement errors results in an ever increasing velocity error (for example, due to sensor noise, biases, scale-factor and alignment errors), which in turn is accumulated into an even more rapidly-increasing position error. The effects of numerical integration error are particularly pronounced in the case of low-cost inertial measurement units (IMUs), the sensors of which are relatively noisy, possess large alignment and scale-factor errors, and have biases which may drift significantly over time or with changes in temperature. The availability of low-cost IMUs is becoming more widespread and, as such, positioning, navigation and tracking systems are becoming more commonplace, for example in smart devices such as smart phones and fitness trackers. However, these low cost IMUs are generally of relatively low quality which means that the error drift is typically substantial, providing spurious and unhelpful results to a user. Global Navigation Satellite System (GNSS) receivers (such as GPS and GLONASS and Galileo) may be used in combination with IMUs in order to help improve the accuracy of the navigation solution. However, GNSS coverage is not always readily available (for example inside buildings) and GNSS units themselves may provide inaccurate navigation data in certain situations, for example in "urban canyons" where tall buildings may block visibility to the satellite constellation. Furthermore, smart devices such as smart phones may be carried in a variety of different positions and orientations with respect to the user, who may frequently change his or her manner of motion. For example, a smart phone user may start a journey by walking with the smart phone positioned in a pocket. On receipt of a call, the phone may be moved to the user's ear, during which time the user has to run for a bus, after which the user spends the remainder of the journey travelling by bus. Such changes in motion of the user and relative position and orientation of the device with respect to the user can introduce further error into the final navigation solution in comparison to, for example, "strapdown" systems (where highly accurate IMUs are provided in fixed positions and orientations with respect to their host body) or gimballed systems. Nevertheless, users of tracking and navigation systems using low cost IMUs still demand accurate and reliable positioning, tracking and navigation information, and there is therefore a requirement in the art to solve the problems outlined above. In particular, there is a requirement to provide accurate positioning, tracking and navigation systems in situations when GNSS positioning data is inaccurate or unavailable. CN104898148 discloses an INS/GPS navigation method based on data compression and a neural network. Gao Nan et al: "An integrated land vehicle navigation system based on context awareness" describes a GPS/micro strapdown inertial navigation system/magnetometer integrated urban navigation system based on context awareness. The integrated system provides estimates of vehicle attitude during GPS outages. Tao Zhang et al: "A new method of seamless land navigation for GPS/INS integrated system" describes a strapdown inertial navigation system(SINS)/GPS/magnetometer integrated navigation system that uses a neural network trained on velocity and position information from the SNIS as input and corresponding errors as output. SUMMARY OF THE INVENTION In accordance with a first aspect of the invention there is provided a computer-implemented method as set out in independent claim 1. The first data are typically obtained by a primary positioning unit mounted on a first platform carried on a body and the second data are typically obtained by one or more secondary sensors mounted on a second platform carried on the body. The second data may be obtained by secondary sensors mounted on a plurality of second platforms carried on the body. The first and second platforms are independent of each other. In other words, they are physically separate and are typically different devices. For example, and as will be explained in more