Patent Application
20100076721
This invention relates to an automated method and system for tracking and dynamically measuring locations in 3-dimensional space to determine one or more dimensions or one or more positions of a body during the performance of a repetitive motion.
Prior art optical-based measurement systems have been employed to measure or analyze motion of a body-including a performance of a repetitious action during some period of time. One example is the Motus system of Vicon Motion Systems (Centennial, Colo.), which employs retro-reflective markers attached to the joints or other locations of a human body and viewed by one or more video cameras. Some systems can track and analyze the motion in 3 dimensions (3-d).
For example, prior art systems may search for a single maximum extension angle of a joint during a motion recording period, and upon conclusion of the recording period, typically report only a one isolated maximum extension angle which was captured. This may be because the system is not specialized for recording and analyzing repetitive—or cyclical—motion. Furthermore, a prior art system may not actually estimate—such as through interpolation—what the actual maximum was, but only the maximum angle of all the body positions which were captured and recorded by the system. That is, motion capture systems acquire and record only discrete body positions, not continuous motion of the body, so that the actual maximum angle in generally may have occurred between two consecutive acquired samples. Other examples include measuring the minimum or flexion angle of a joint, the angle of the joint at some point in a repetitive stroke, a minimum or maximum distance between points on a body, or a distance between body points based on the recorded point locations acquired at discrete instants in time.
Described herein are a system, an apparatus, and a method, among other embodiments, adapted to obtain dynamic sizing measurements. As used herein, the term “dynamic sizing measurements” refers to one or more body dimensions taken during the performance of a repetitive—or cyclic—action. Dynamic sizing measurements may be used in a variety of applications. One application that employs dynamic sizing measurements is the fitting of sporting equipment to specific users. One type of sporting equipment which may use dynamic sizing measurements to properly fit the equipment to a specific user is a bicycle. It is to be appreciated that the systems, apparatus, methods, and other embodiments described herein may be applied to other sporting equipment and non-sporting equipment. Furthermore, the systems, apparatus, methods, and other embodiments described herein may be applied in non-fitting applications such as, but not limited to, other biomechanical or healthcare applications.
One embodiment comprises a method of taking measurements of a cyclist 1 situated on a bicycle 2 while the cyclist 1 is operating the bicycle 2 in a stationary position—such, as, but not limited to, on a trainer 3, as shown in
In a method called Stroke Intelligence, the method—or an apparatus or system implementing the method—determines at least one dimensional statistic such as an average minimum and/or an average maximum angle or distance over a plurality of strokes—or cycles—of repetitive motion. The minimum or maximum dimensional statistic is not based on a single measurement location at a single instant of time within a single stroke, but the dimensional statistic is really an average of the minimal or an average of the maximal dimensions computed from the marker locations determined during a plurality of the strokes of the repetitive motion.
The motion may be represented by a sequence of coordinates and corresponding timestamps for each marker, where the coordinates represent the locations, and where the timestamps represent the instants in time when the locations were determined.
The accompanying drawings, which are incorporated herein and form a part of the specification, illustrate a preferred embodiment of the present invention and, together with the description, serve to explain the principle of the invention.
Table 1 lists various possible computed measurements using locations of markers measured by the system.
One embodiment uses an optical measurement system. With respect to
In one embodiment, each marker 10a-10w is adapted to sequentially flash and emit light. For example, as shown in
In one embodiment, it may be advantageous to calculate a maximum knee extension angle in order to properly fit a bicycle 2 to a cyclist 1. In order to calculate an accurate maximum knee extension angle 40, the maximum knee extension angle 40 is calculated for each stroke in a series of consecutive strokes, wherein, a single stroke—or cycle of motion may be characterized as a complete revolution of a pedal crank 5. In determining one knee extension angle 40, the first marker 10h may emit light for 3.5 ms at a first time, the second marker 10k may emit light for 3.5ms at a second time, wherein the second time follows the first time, and the third marker 10a may emit light for 3.5 ms at a third time, the third time immediately following the second time. Longer or shorter light emitting periods may be used, the markers may emit light in some other order, and/or additional markers 10f, 10s, 10w may sequentially emit light. In one embodiment, the location of these markers 10a-10w during a stroke is digitized by the reception unit 22 and then a signal characterizing a location for each marker 10a-10w is sent via a cable 23 to a processing unit 24. One processing unit 24 may be a laptop, some other personal computer, or a stand-alone embedded computer. The processing unit 24 is adapted to acquire the marker location data received from the reception unit 22 and process the acquired data into 3-dimensional coordinate values. These 3-d coordinate values may then be used for further processing and data manipulation by the processing unit or a separate computer.
In a process called “stroke intelligence”, further explained below, one embodiment may take sets of measurements from each stroke and average together the corresponding measurements. In one embodiment, a system may have knowledge of the expected repetitive movements of the cyclist and thus can respond to specific key measurement positions of the cyclist. For example, a pedal and a foot coupled to the pedal may generally follow an approximately circular pattern as diagrammed in
Furthermore, the locations of all the markers 10a-10w may be estimated for one and the same specific instant in time using the same technique. Then, the angle formed by any three of the markers 10a-10w will be, in effect, determined accurately from the three locations for the same given instant in time.
All motion detection devices, 3D and video, have a set acquisition frequency and therefore do not capture all points continuously. One feature of the present invention is the ability to obtain sufficiently accurate estimates of locations, distances, and/or angles even when the reception unit 22 does not capture marker locations at the optimal time within a given stroke. The estimates may be reliably obtained through software interpolation based on a set of measurements acquired before and after the optimal time. For example, due to the known application-specific movements of the cyclist, such as the foot being attached to the pedal, and the pedal being attached to the crank 5, the foot is known to move approximately in a circle and therefore, more accurate foot marker locations may be estimated. Other body parts may repetitively move along 3-d geometrical curves other than a circle. This system of interpolation may be known as “Stroke Intelligence”.
Instead of a circle, some other curve, such as a parabola or other polynomial may be used to approximate the continuous 3-d path of a marker and estimate an extreme location or to estimate the location of the marker at some specified moment.
In some embodiments, it may be preferred to obtain more than just the locations of markers on a cyclist's body parts. For example, collections of body positions that make up cyclically-changing angles are desired. A maximum or minimum of such an angle for each stroke may be estimated, such as the angle formed by markers 10h, 10k, 10a, which may represent the knee angle 40 formed by the thigh and calf. The maximum angles formed by marker locations—as calculated above—spanning a plurality of strokes may be averaged together. That is, one maximum angle may be estimated for each of the plurality of repetitive strokes. Then the average of the estimated maximum angles may be used as a substantially reliable and accurate measurement of the cyclist's knee extension angle. Similarly, the estimated minimum angles for all strokes may be averaged together to provide a substantially accurate measurement of the cyclist's knee flexion angle. Likewise, the measurement of an angle or a distance at a given point in each repetitive stroke may be combined with the corresponding measurements of all other strokes to form an average or consensus value.
Further dimensional statistics besides average minima or average maxima—such as ranges, means and standard deviations of locations, distance, or angles—may be collected over a period of time. The statistics may be collected for any or all angles defined by three markers or for any or all distances between two given markers. Statistics may be gathered similarly for other measureable, dimensional attributes, such as area, volume, power output, or speed.
An example of Stroke Intelligence computation is measuring the knee extension angle 40. Nevertheless, as shown in Table 1 and
Distance dimensions as well as angles may be estimated using Stroke Intelligence, and dimensional statistics may be computed therefrom. For example, it may be useful to measure the horizontal distance of the foot with respect to the knee when the foot is at the most-forward position. That is when the pedal crank is at the “3 o'clock” angle for the right side of the cyclist, or at the “9 o'clock” angle for the left side. Few, if any, of the locations of the foot marker may have been acquired with the foot exactly in this location. However, stoke intelligence can use three or more foot locations 81a,81b,81c of the marker 10f to estimate when and where the foot marker 81c would have reached its most forward location 82 during each stroke by using non-linear circular or polynomial functions to estimate the minimum or maximum of the function. Finding a minimum or maximum of a function is a well known method in elementary calculus.
Incorporated into the calculations is “marker intelligence”. This means that the system knows which marker is which. In other words, the system knows that light received by the reception unit 20 at a certain instant in time applies to a specific marker 10. In prior art video systems, a video system operator would have to manually seek each marker and calculate the desired measurement for each stroke and then average the measurements together. The prior art method ignores the problem of interpolation when no captured video frame aligns with the desired cyclist position. Further inaccuracy is introduced by the unreliability of manually selecting the desired markers repeatably on a small computer screen.
The description above has assumed that the locations—specifically the location coordinates—of the markers and the measurements based on the locations are within a 3-dimensional space. The system, apparatus, and method can be equally applied to locations and measurements within a 2-dimensional space.
Those skilled in the art can readily recognize that numerous variations and substitutions may be made in the invention, its use, and its configuration to achieve substantially the same results as achieved by the embodiments described herein. Accordingly, there is no intention to limit the invention to the disclosed exemplary forms. Many other variations, modifications, and alternative constructions fall within the scope and spirit of the disclosed invention as expressed in the claims.
This a US non-provisional patent application claiming priority to the provisional patent application filed by Simms, et al, on Sep. 23, 2008, with Ser. No. 61/099,490.
| Number | Date | Country | |
|---|---|---|---|
| 61099490 | Sep 2008 | US |
