GB2400686A

GB2400686A - Motion logging and robotic control and display system

Info

Publication number: GB2400686A
Application number: GB0307795A
Authority: GB
Inventors: Christopher Charles Box
Original assignee: Individual
Current assignee: Individual
Priority date: 2003-04-04
Filing date: 2003-04-04
Publication date: 2004-10-20
Also published as: GB0307795D0

Abstract

A bipedal robot 21 has articulated legs 23L, 23R moved by servo motors 29L, 29R, 30L, 30R, 31L, 31 R at its simulated hip, knee and ankle joints, respectively. These are controlled by an electronic control system 25 which is fed motion data by a remote computer, along wires or wirelessly. The motion data may be captured using a harness 1 worn by an experimental subject 2. The harness 1 is strapped to each leg of the subject 2 and has potentiometers at its ankle 9, knee 10 and hip joints 11, aligned with the subject's corresponding joints, which provide data on angular displacement of the joints against time. These data may be transmitted to the robot 21 in real time, or stored for later transmission, possibly after further processing. The computer may also display the data on-screen as graphs and/or as a mannequin image.

Description

MOTION LOGGING AND ROBOTIC DISPLAY SYSTEM

The present invention relates to a system to capture, record and display data concerning human body movement, such as limb movements during walking or running. More particularly but not exclusively, it relates to such a system comprising a humanoid robot controllable to display said data.

"Motion capture" systems are known which can detect and record electronically the body movements of a subject. For example, the movements of a footballer or other sportsman have been recorded and used to enhance the realistic appearance of a corresponding moving image In a computer game.

Such known systems are extremely elaborate and expensive to set up and to operate. In one system, the subject has a plurality of reflective markers attached to his or her person, and is monitored by a high-resolution stereoscopic video system, connected to a computer capable of calculating and logging the three-dimensional position of each marker by combining individual camera images. This requires expensive equipment and very accurate positioning of the individual cameras relative to one another. It is also preferable that the subject wears black or very dark clothing, performs against a black or very dark background, and performs under very bright lights, to maximise the visibility of the markers.

Another drawback is that while each camera may be able to pan to follow the subject, it must maintain its overall position relative to the other camera(s) i.e. it cannot track. It is therefore very possible that in the course of normal unrestrained movement, a marker may become obscured from one or more cameras. Unless a marker is in view from at least two cameras, its position cannot accurately be followed.

Another known motion capture system comprises a specially constructed chamber equipped with a plurality of sensors capable of tracking magnetic markers attached to a subject's body. Although obscuration of such markers may not be a problem, equipping and setting up such a chamber is believed to be even more expensive than the stereoscope arrangement described above.

Neither arrangement is particularly suitable for use by small and medium sized enterprises, in education or by private enthusiasts or clubs. Neither arrangement can conveniently be taken to a desired venue - the subject must normally go to an established facility where the system is already set up.

It would be desirable, therefore, if a system were available which could capture, record and display bodily movements, and which was sufficiently economical to produce and simple to operate to be suitable for use in education, by small and medium sized businesses, by sports clubs or even by individual enthusiasts. Such a system should also be usable "in the field" with a minimum of ancillary equipment.

For the majority of applications, movements of minor joints, such as those of the fingers and toes, are of little importance. A suitable system would therefore focus on movements of the hip, knee and ankle joints of the leg (hereinafter referred to as "major leg joints") and the shoulder, elbow and wrist joints of the arm (hereinafter referred to as "major arm joints"). Together, these joints will be referred to as "major limb joints" of the body.

One field in which such a system could be particularly applicable is robotics. There has been considerable development associated with arm movements, spurred for example by the requirements of equipment for assembly lines. Robotic "arms" are now available that can simulate or even improve on the range of movements possible for a human arm.

Such arms are, however, usually elaborate pieces of equipment on a fixed mounting.

Robotic locomotion has of late concentrated mainly on wheeled machines. There has been some work on "walker" robots having six, eight or more legs and imitating the gait of insects, arachnids and centipedes and the like. However, effective bipedal locomotion has proven difficult to address, particularly with legs having as many independent joints as a human leg. Not only is balance a practical problem, but theoretical dynamic analyses of how a pair of legs should move have not proven as useful as might be expected. Getting a bipedal robot to "walk" is hard enough, let alone achieving "running" or more extreme leg movements, such as in many sports.

It is believed that the provision of data on authentic human leg movements may be key to robotic simulation of walking, running or the like. A robot capable of moving in response to such data would also be a superior means of displaying and comparing gait data from different subjects, for example.

It is therefore an object of the present invention to provide a system for recording and displaying bodily movements which is simpler and more widely applicable than existing systems. It is another object of the present invention to provide such a system comprising mechanical display means for the recorded data. It is also an object of the present invention to provide a method of operation of such a system, and to provide mechanical display means for such a system.

According to a first aspect of the present invention, there is provided motion capture apparatus comprising an articulated harness mountable to at least one of the limbs of a subject and provided with a plurality of means for detecting angular displacements, and data logging means connectible thereto, wherein the harness comprises joints alienable with each of the major joints of a limb of the subject and each joint of the harness comprises at least one said angular detection means.

Preferably, said at least one limb of the subject comprises a leg, and said major limb joints then comprise a hip, a knee and an ankle of the leg.

Alternatively or additionally, said at least one limb comprises an arm of the subject, and said major limb joints then comprise a shoulder, an elbow and a wrist of the arm.

In a preferred embodiment, the harness is mountable to both legs of the subject.

Preferably, the harness comprises a pair of first elongate connecting members each extending between a joint alienable with a hip and a respective joint alienable with a knee, and a pair of second elongate connecting members each extending between a joint alienable with a knee and a respective joint alignable with an ankle.

Advantageously, the harness comprises a pair of members, each of which is attachable to a respective foot of the subject and is connected to a respective joint alienable with an ankle.

Optionally, the harness comprises belt means fastenable around a torso of the subject and a pair of third elongate connecting members, each of which extends between the belt means and a respective joint alienable with a hip.

Each first and second connecting member may be provided with means to fasten it to an adjacent part of the subject's leg, preferably by adjustable fastening means.

Each first and second elongate connecting member may be so longitudinally adjustable that a desired separation between a respective pair of joints of the harness may be selected.

Each third elongate connecting member may be so mounted to the belt means that a desired separation between the belt means and each joint alignable with a hip can be selected.

The harness may thus be adjusted such that each joint thereof is in alignment with a t corresponding joint of the leg of the subject.

Each angular detection means preferably comprises potentiometer means connectable via respective electric cable means to the data logging device.

The data logging means may comprise computing means, preferably provided with means to display said data.

In this case said display means comprises display screen means.

The computing means may then be programmed to display the data in the form of graphs of angular displacement of any or each angular detection means against time.

The computing means may alternatively or additionally be programmed to display the t data in the form of an image representing a leg of the subject with the joints of the legs of i the image shown at angles corresponding to the data.

The computing means may be so programmed that a user may compare said data with further sets of data, such as data taken previously for the same subject or for other subjects.

Alternatively or additionally, the display means may comprise controllable robot means, for example controllable bipedal robot means. t Said bipedal robot means may comprise a robotic device as described in the third aspect below.

According to a second aspect of the present invention, there in provided a motion capture system comprising motion capture apparatus adapted to collect data representative of angular displacements of joints of an experimental subject's body at preselected intervals of time and computing means programmed to record said data and to display a representation thereof on display means of the system.

Preferably, said joints of the body comprise major limb joints thereof, optionally major leg joints thereof.

Said display means may comprise display screen means.

The computing means may then be programmed to display the data in the form of angular displacement of any or each joint of the subject's body against time.

The computing means may alternatively or additionally be programmed to display the data in the form of an image representing at least one of the limbs of the subject with joints of the or each limb, for example both legs, of the image shown at angles corresponding to the data.

The computing means may be programmed to allow electronic manipulation of the data collected.

Said manipulation may comprise comparing the data with data taken previously and/or for other experimental subjects.

Said manipulation may comprise comparing the data with data produced from theoretical models.

Said manipulation may comprise selecting a portion of the data, for example corresponding to a step taken by the subject, and generating a further set of data from repetitions of said portion.

Alternatively or additionally, the display means may comprise controllable robot means, such as a bipedal robotic device as described in the third aspect below.

According to a third aspect of the present invention, there is provided a robotic device adapted for bipedal locomotion comprising a pair of articulated leg means, each leg means comprising hip joint means, knee joint means and ankle joint means and respective motor means adapted to move each said joint means in response to commands from control means of the device.

Preferably, said control means comprises electronic data processing means mounted to the device, optionally to a torso section thereof to which the leg means are mounted.

Advantageously, said data processing means sends commands to the motor means in response to data transmitted to the device from remote computing means.

Said data may be transmitted by electric cable means.

Alternatively, it may be transmitted to the device by wireless communication means.

The data transmitted to the device preferably comprise data collected from a motion capture apparatus, said data comprising angular displacements of joints of the legs of an experimental subject.

Said data collected from the motion capture apparatus may be modified by the computing means before transmission to the device.

Preferably, each motor means comprises an electrically powered servo motor adapted to move a respective joint means pivotingly to a selected angle.

Optionally, any or each joint means may be pivotable about more than one axis.

The device may be supplied with power for the servo motors along electrical cable means connected to a remote power source.

Alternatively, the robotic device may be provided with an integral power source.

In a first embodiment, the robotic device is mounted to support means.

Preferably, the robotic device is so mounted to a stationary support that it may move relative thereto, for example in circles around the support.

Alternatively, the support means may move with the robotic device.

In a second embodiment, the robotic device is provided with means to maintain its balance.

Said balancing means preferably comprises sensor means adapted to detect any toppling movement of the device away from a substantially vertical orientation.

The sensor means may comprise pendulum means and/or gyroscope means.

Advantageously, said sensor means is adapted to detect such toppling movement in more than one plane.

Either the remote computing means or the control means of the device may then modify the commands sent to the motor means to produce movements of the leg means to counteract or compensate for the toppling movement of the device.

According to a fourth aspect of the present invention, there is provided a method for capturing data concerning limb movements of an experimental subject comprising the steps of providing motion capture apparatus as described in the first aspect above, so attaching it to the person of the subject that each joint of the motion capture apparatus is substantially aligned with a corresponding major joint of a limb of the subject, causing the subject to perform an action such as walking, and recording at preselected intervals data on the angular displacement of each said joint of the apparatus.

According to a fifth aspect of the present invention, there is provided a method of controlling the motion of a bipedal robotic device comprising the steps of recording at preselected intervals data concerning angular displacement of the joints of the legs of a living subject and commanding each corresponding joint of the robotic device to adopt the same angular displacement at a corresponding time interval.

Preferably, the device is a robotic device as described in the third aspect above.

Advantageously, the method comprises the steps of capturing data as described in the fourth aspect above.

Embodiments of the present invention will now be more particularly described by way of example and with reference to the accompanying drawings, in which: Figure 1 is a perspective view of motion capture apparatus of the invention being Figure 2 is a screenshot of a computer program of the invention displaying data from the apparatus of Figure 1; Figures 3A to 3C are graphs of data from hip, knee and ankle sensors of the apparatus of Figure 1, corresponding to the screenshot of Figure 2; Figure 4 is a frontal elevation of a walking device of the invention; Figure S is a perspective view of the walking device of Figure 4 in action, mounted to a counterbalanced support; and Figure 6 is a perspective view of the walking device of Figure 4 in action, connected in real time to the motion capture apparatus of Figure 1.

Referring now to the Figures, and to Figure 1 in particular, a motion capture apparatus 1 is shown being worn by an experimental subject 2. The motion capture apparatus 1 comprises a waist belt section 3 to which are mounted two separate leg units, each of which is substantially a mirror image of the other. (Only a left leg unit 4 is visible in this Figure).

The left leg unit 4 comprises a pelvic section 5, a thigh section 6, a lower leg section 7 and a foot section 8. The sections 5, 6, 7, 8 are each jointed pivotably to their neighbour section or sections; the foot section 8 is connected via an ankle joint 9 to the lower leg section 7; the lower leg section 7 is connected via a knee joint 10 to the thigh section 6; and the thigh section 6 is connected via a hip joint 11 to the pelvic section 5. The pelvic section 5 is connected via a sliding fitting 12 to the waist belt section 3. i The pelvic section 5 comprises an elongate strip of semi-rigid plastics material. Each of the thigh section 6 and the lower leg section 7 comprises two overlapping elongate strips of semi-rigid plastics material, mounted adjustably one to the other to form a single elongate member of selectable length. For example, a pair of threaded studs spaced longitudinally along a first said strip may extend through a longitudinal slot in a second said strip so that the strips may be moved slidingly longitudinally one relative to the other. Once a selected overall length is achieved, a nut on each stud is tightened to fasten the strips together.

The foot section 8 is provided with a plastics plate 13 which is formed to fit over an instep of a subject's foot.

Adjustable straps 14 are provided to hold each of the foot section 8, lower leg section 7 and thigh section 6 to a corresponding part of the subject's own leg. Conveniently, the straps 14 are provided with corresponding patches of hook and loop repositionable fastening material so that they can be fastened together at a required length. They may also be elasticated for comfort.

Each of the ankle joint 9, knee joint 10 and hip joint 11 comprises a potentiometer, which detects the angle to which the joint is pivoted. Cabling 15 leads from each potentiometer to a data transmission unit, which passes this angular data to a logging computer, either i along a cable or more conveniently via a wireless data link.

Thus, to use the motion capture apparatus 1, the subject 2 fastens the waist belt section 3 around his waist, and straps each of the right and left leg units 4 to an outside of his corresponding leg, with the plate 13 of each foot section 8 firmly in place on the instep of a corresponding foot. The lengths of each thigh 6 and lower leg section 7 are adjusted, and the pelvic section 5 is slid within the fitting 12 on the waist belt section 3, until each of the ankle 9, knee 10 and hip 11 joints of the apparatus 1 is exactly in line with an axis of rotation of a corresponding joint of the body of the subject 2. The subject 2 is then free to move normally, and the apparatus 1 will follow his leg movements without hindering them to any significant extent.

Data capture is preferably started with the subject 2 standing bolt upright, to establish a reference zero point for each potentiometer. The angular position of each potentiometer is then transmitted, fifty times per second, to the computer, where it is logged.

The computer is programmed to collect this data as a time series and to display it in various formats. For example, as shown in Figure 2, separate graphical displays 16, 17, 18 of angle against time can be shown for each right and left leg pair of potentiometers.

It is convenient to display a corresponding right and left leg potentiometer trace on a single graph for comparison purposes. Data is also presented in a numerical display 19 of an angle of each potentiometer at any given time, and in the form of a mannequin image whose hip, knee and ankle positions track those of the corresponding potentiometers i of the apparatus 1. (The arm positions of the manikin 20 are purely decorative in the display shown, although the system described could also be adapted to capture and display upper limb motions).

The graphical displays 16, 17, 18, showing hip, knee and ankle angular displacement, respectively, against time, are shown in more detail in Figures 3A to 3C respectively.

In each of Figures 3A to 3C, the y-axis of the graph represents an angular displacement, in degrees, relative to a bolt-upright standing position as zero. The x-axis of each graph represents time; each of Figures 3A to 3C is on the same time scale for ease of comparison. Traces for corresponding right and left potentiometers are superimposed, marked with "R" and "L" to help distinguish them.

These graphs represent a typical sequence of steps for a walking subject. As can be seen, none of the joints moves as regularly as might be expected from mathematically modelling a leg as a series of pendula. The hip joint graph 16 is closest to sinusoidal, but there can be considerable variation between successive steps. The knee joint graph 17 shows even more variation between successive steps and shows many additional peaks in what appears to be a broadly sinusoidal trace. The shape of the ankle joint graph 18 is substantially more complex than a simple model would predict.

It is thus evident that the motion capture apparatus 1 can provide considerable amounts of detailed data, even on a few steps of a walking gait. It has been found that there is considerable variation between individuals, in both the form of the graphs 16, 17, 18 and the variation between successive steps within each graph.

There are many potential uses for such data. For example, a coach could collect data on an athlete, and use it to identify existing or potential shortcomings, perhaps by comparing the athlete's results to those of a more successful athlete, or to an averaged set of data combining results from many subjects. Clearly, additional displays could show the differences between an individual's results and an ideal or generic data set, to highlight discrepancies.

Such an approach could also be of use medically, allowing a physiotherapist or the like to diagnose and hence treat a problem based on unusual features of a particular joint's movements. Differences between the graphs for a subject's corresponding right and left joint could be of particular utility in this regard.

In either use, changes in a subject's results over the course of training or therapy could be analysed and displayed appropriately.

Another application envisaged for such data is in cartoon animation. Much of the effect of caricature is believed to stem from exaggerating or emphasising where the subject differs from the norm. Thus, comparing an individual's data with an averaged set, and emphasising the differences, could allow the creation of a cartoon figure with a gait allowing rapid recognition of that individual. Also, such differences could be applied to ideal motion data sets for animal or birds, for example, allowing the creation of a cartoon creature with some of the "body language" of a famous human.

One area in which data collected by the apparatus 1 would be of particular applicability is robotics, and especially the problem of bipedal locomotion. As mentioned above, simple theoretical models of walking are not sufficient for the production of an effective mechanical bipedal gait. However, a robot programmed to walk according to a genuine human gait is found to be much more successful. Such robots would also be useful as research tools or as educational aids.

A bipedal robot 21, which can operate on the basis of motion capture data from the apparatus 1, is shown in Figure 4. The robot 21 comprises a torso section 22 supported by a right leg section 23R and a matching left leg section 23L. The torso section 22 comprises a back plate 24, to which is mounted a control system 25 comprising logic controllers, electrical power supply components and associated electronic equipment, which powers and controls the leg sections 23R, 23L.

Each leg section 23R, 23L comprises a thigh element 26R, 26L, a lower leg element 27R, 27L and a foot element 28R, 28L. The thigh element 26R, 26L of each leg section 23R, 23L is pivotably connected to the torso section 22 by means of a respective hip servo motor 29R, 29L. The lower leg elements 27R, 27L are each pivotably connected to a corresponding thigh element 26R, 26L via a respective knee servo motor 30R, SOL, and the foot elements 28R, 28L, are each pivotably connected to a corresponding lower leg element 27R, 27L via a respective ankle servo motor 31 R. 31 L. Each servo motor 29R, 29L, 30R, 30L, 31R, 31L is independently controlled and supplied with electrical power by the control system 25. Data collected from the potentiometer on each joint 9, 10, 11 of the motion capture apparatus 1 is supplied to the control system 25, which transforms it into instructions for corresponding angular movements of each servo motor 28R, 29L, 30R, 30L, 31R, 31L. The leg sections 23R 23L of the robot 21 will thus execute walking motions, etc. corresponding to those of the experimental subject 2 wearing the motion capture apparatus 1.

The robot 21 can thus "walk" following the subject's 2 natural gait as displayed in Figures 2 and 3. Alternatively, the computer logging and analysing data from the apparatus 1 can be programmed or controlled to select a single cycle of data, corresponding to one step with each leg by the subject 2, and to repeat that cycle to form a standardised gait. An average of several steps could also be used as the "model" step of a standardised gait. It is also possible to supply the robot 21 with motion data from theoretical models, its resulting movements providing experimental evaluation of the validity or usefulness of the theoretical models.

The robot 21 shown in Figure 4 is a basic version, requiring an electrical supply from a remote source, for example via a power cable. Other versions may be provided with in- built power supplies, although the weight of a battery or the like of sufficient capacity may require that it is incorporated in a trailer or the like, rather than mounted to the robot 21 itself (which might harm its stability).

The robot 21 shown is also intended to be provided with motion control data along a permanently connected cable. Other versions, particularly those with autonomous power supplies, could be provided with wireless telemetry links, or have the required data provided on pre-programmed replaceable data chips or the like.

Although it "walks" using authentic human motion data, the robot 21 shown cannot react e.g. to unevenness in a surface on which it is walking in the way that a human does. It is also operating with power and data cables dragging behind it. Both factors can result in balance problems.

It is therefore operated mounted to a stand, as shown in Figure 5. The torso section 24 of the robot 21 is connected at around its "shoulder" region to a boom 32, which is mounted adjacent its midpoint to a stand 33. An adjustable counterweight 34 is mounted to an end of the boom 32 remote from the robot 21. The boom 32 and the stand 33 are connected by a rotatable joint 35.

Power and data are transmitted along cables (not shown) to a connector unit 36 mounted to the stand 33, and thence through slip ring electrical contacts within the rotatable joint and cabling within the boom 32 to the control system 25 of the robot 21.

The robot 21 thus "walks" or "runs" in a circle around the stand 33, relieved of the weight of its power and data cables, and is prevented from falling over by the counterweight 34 at the remote end of the boom 32.

As well as operating in response to recorded, manipulated or modelled data, the robot 21 can also be supplied with data from the motion capture apparatus l in real time, as shown in Figure 6. The subject 2 wearing the apparatus l is walking along, while the robot 21, mounted to the stand 33, follows his leg movements precisely. The subject 2 and robot 21 need not, of course, be present together.

For a robot capable of autonomous movement, self-balancing and able to change direction, the leg joints described above, containing simple servo motors 29R, 29L, 30R, SOL, SIR, 31L capable of controlled pivoting motion about a single axis, are replaced by joints with added degrees of motion to allow fully articulated movement. The human hipjoint is a ball-andsocket joint capable of much more than simple pivotal movements, and the ankle joint also allows the foot to be rotated about more than one axis simultaneously. These increased degrees of freedom allow small corrections and alterations in the motion of the leg to aid in balancing, for example.

Thus, such a robot would be supplied with motion data combining a sequence of captured steps, or a repeated selected, averaged or idealised step, with corrections to allow it to balance or turn. The robot would be provided with balance sensors operating in both a fore-and-aft plane and a side-to-side plane, to detect and report if it is losing its balance. (These sensors could be based on pendula, gyroscopes or solid-stategyroscope devices). The computer sending motion data to the robot would modify these data in real time so that the robot's movements react to and compensate for any impending loss of balance.

Such an arrangement would also allow a user to control the robot, for example with a joystick connected to the computer. The authentic captured walking (or running, etc) data being transmitted to the robot would be modified by the computer to make the robot turn to one side or the other as it moves along. Different "model" steps could also be kept in memory associated with different speeds, to be selected by a user-operated controller, so that the robot could shuffle, walk, pace, march, stride, jog, run or even sprint depending on the speed required.

It is envisaged that motion capture apparatus of the types described above may have several further applications. For example, data captured from a motion capture apparatus mounted to a remaining limb of an amputee could be used to control a prosthetic apparatus replacing an amputated limb, either directly or by using said data to create "model" movements for the prosthetic apparatus.

Similarly, a stroke victim may have impaired motion in one side of his or her body, while the other remains fully under nervous control. Thus, data could be collected from a healthy limb and used to control (or assist in the control of) movements of the corresponding impaired limb.

Paraplegics could be aided in walking using data captured from both arms and legs of a healthy subject. The arm and leg motions of the subject would be correlated (e.g. arms are normally swung exactly out of phase with leg movements during walking, etc). The paraplegic would wear an arm harness (such as the motion capture device of the invention) detecting his arm movements, and these arm movements would be used to determine appropriate leg movements, via a computer-based "look-up table" or the like.

Commands would then be issued to a powered leg harness worn by the paraplegic, which would move his or her legs accordingly.

A variant on this concept could also be used to allow a quadriplegic to move. A powered limb harness worn by the quadriplegic could receive "realtime" commands derived from a motion capture apparatus worn by a helper.

Alternatively, the quadriplegic could be provided with a powered harness controlled using recorded motion capture data. The quadriplegic would be provided with a control device, operable for example with his or her mouth, or with eye movements, and an artificial intelligence package could select appropriate recorded movements for transmission to the powered harness, based on commands given via the control device.

A motion capture harness such as those described above could be incorporated into a system for controlling movements of paralysed limbs by direct electrical stimulation of the muscles of the limb, to provide a simple feedback mechanism. Currently in such systems, muscles are contracted and relaxed using powerful electric pulses which cause only maximum contraction or relaxation. This produces sudden, exaggerated movements and rapid muscular fatigue. Using motion capture data from such a harness, the electric pulses passed to the muscles can be reduced to whatever level is necessary to maintain a limb in a desired position or cause a desired movement, and no more. This would reduce fatigue, increase control of movements, reduce the risk of tearing muscular tissue from overstimulation and increase the period for which the paralysed limb could be used.

It is also envisaged that data could be mapped from leg joints to corresponding arm Joints, or vice versa.

A desirable variant of the harness would comprise sensors fixed to major limb joints to move therewith, connected only by wiring, and without the harness members, connecting the sensors at each joint, used in the basic harnesses described above.

A waterproofed version of any of the harnesses described would be of particular use to capture data from a swimming subject, or one engaged in outdoor pursuits, such as climbing.

Claims

Claims 1. A robotic device adapted for bipedal locomotion, comprising

control means, a pair of articulated leg means, each leg means comprising hip joint means, knee joint means and ankle joint means and respective motor means adapted to move each said joint means in response to commands from said control means.
2. A robotic device as claimed in Claim 1, further comprising a torso section to which the leg means are mounted.
3. A robotic device as claimed in either claim I or claim 2, wherein said control means comprises electronic data processing means mounted to the device.
4. A robotic device as claimed in claim 3, wherein said data processing means is adapted to send commands to the motor means in response to data transmitted to the device from remote computing means.
5. A robotic device as claimed in Claim 4, wherein the data transmitted to the device comprise angular displacements of joints ol the legs of an experimental subject collected lrom a motion capture apparatus.
6. A robotic device as claimed in Claim 5, wherein the remote computing means is adapted to modify said data collected from the motion capture apparatus before transmission to the device.
7. A robotic device as claimed in any one of Claims I to 6, wherein each motor means comprises an electrically powered servo motor adapted to move a respective joint means pivotingly through a selected angle.
8. A robotic device as claimed in any one of Claims 1 to 7, wherein one or more of said joint means is pivotable about more than one axis.
9. A robotic device as claimed in any one of Claims I to 8, mounted to support means.
10. A robotic device as claimed in Claim 9, so mounted to a stationary support that it may move relative thereto, for example in circles around the support.
11. A robotic device as claimed in Claim 9, wherein the support means moves with the robotic device. 1
12. A robotic device as claimed in any one of Claims I to 8, provided with means to maintain its balance.
13. A robotic device as claimed in Claim 12, wherein said balancing means comprises sensor means adapted to detect any toppling movement of the device away from a substantially vertical orientation, such as pendulum means and/or gyroscope means.
14. A robotic device as claimed in Claim 13, wherein the remote computing means is adapted to modify the commands sent to the motor means to produce such movements of the leg means as to counteract or compensate for the toppling movement of the device.
15. A robotic device as claimed in Claim 13, wherein the control means is adapted to modify the commands sent to the motor means to produce such movements of the leg means as to counteract or compensate for the toppling movement of the device.
16. A robotic device substantially as described herein and with reference to the Figures of the accompanying drawings.
17. Motion capture apparatus comprising an articulated harness mountable to at least one limb of a subject and provided with a plurality of joints each alignable with 1 one of the major joints of said at least one limb, with each joint of the harness having at least one said angular displacement detection means, and data logging means connectible thereto.
18. Motion capture apparatus as claimed in Claim 17, wherein said at least one limb of I the sub ject comprises a leg, and said major limb joints comprise a hip, a knee and an ankle of the leg.
19. Motion capture apparatus as claimed in either Claim 17 or Claim 18, wherein said at least one limb comprises an arm of the subject, and said major limb joints comprise a shoulder, an elbow and a wrist of the arm.
20. Motion capture apparatus as claimed in any one of Claims 17 to 19, wherein the harness is mountable to both legs of the subject.
21. Motion capture apparatus as claimed in Claim 20, wherein the harness comprises a pair of first elongate connecting members each extending between a joint alignable with a hip and a respective joint alignable with a knee, and a pair of second elongate connecting members each extending between a joint alignable with a knee and a respective joint alignable with an ankle.
22. Motion capture apparatus as clawed in either Claim 20 or Claim 21, wherein the harness comprises a pair of members, each of which is attachable to a respective toot of the subject and is connected to a respective joint alignable with an ankle.
23. Motion capture apparatus as claimed in any one of Claims 21 to 22, wherein each first and second connecting member is provided with means to fasten it to an ad jacent part of the subject's leg, preferably by adjustable fastening means.
24. Motion capture apparatus as claimed in any one of Claims 20 to 23, wherein the harness is adjustable such that each joint thereof is in alignment with a corresponding joint of the leg of the subject.
25. Motion capture apparatus as claimed in any one of Claims 17 to 24, wherein each angular detection means comprises potentiometer means connectable via respective electric cable means to the data logging means.
26. Motion capture apparatus as claimed in any one of Claims 17 to 25, wherein the data logging means comprises computing means, provided with means to display said data.
27. Motion capture apparatus as claimed in Claim 26, wherein the display means comprises controllable robot means, for example a robotic device as claimed in any one ofClaims I to 16.
28. Motion capture apparatus substantially as described herein and with reference to the Figures ofthe accompanying drawings.
29. A motion capture system comprising motion capture apparatus adapted to collect data representative of angular displacements of joints of an experimental subject's body at preselected intervals of time and computing means programmed to record said data and to display a representation thereof on display means of the system.
30. A motion capture system as claimed in Claim 29, wherein said joints of the body comprise major limb joints thereof, optionally major leg joints thereof.
31. A motion capture system as claimed in either Claim 29 or Claim 30, wherein said display means comprises display screen means.
32. A motion capture system as claimed in Claim 31, wherein the computing means is programmed to display the data in the form of graphs of angular displacement of any or each joint of the subject's body against time.
33. A motion capture system as claimed in either Claim 31 or Claim 32, wherein the computing means is programmed to display the data in the form of an image representing at least one of the limbs of the subject with joints of the or each limb, for example both legs, of the image shown at angles corresponding to the data.
34. A motion capture system as claimed in any one of Claims 29 to 33, wherein the computing means is programmed to allow electronic manipulation of the data collected.
35. A motion capture system as claimed in Claim 34, wherein said manipulation comprises comparing the data with data taken previously and/or for other experimental subjects, and/or comparing the data with data produced from theoretical models.
36. A motion capture system as claimed in either Claim 34 or Claim 35, wherein said manipulation comprises selecting a portion of the data, for example corresponding to a step taken by the subject, and generating a further set of data from repetitions -I of said portion.
37. A motion capture system as claimed in any one of Claims 29 to 36, wherein the display means comprises controllable robot means, such as a bipedal robotic device as claimed in any one of Claims I to 16.
38. A motion capture system substantially as described herein and with reference to the Figures of accompanying drawings.
39. A method for capturing data concerning limb movements of an experimental subject, comprising the steps of providing motion capture apparatus as claimed in any one ol Claims 17 to 28, so attaching it to the person ol the subject that each joint of the motion capture apparatus is substantially aligned with a corresponding major joint of a Ihnb of the subject, causing the subject to perform an action such as walking, and recording at predetermined intervals data on the angular displacement of each said joint of the apparatus. I
40. A method for capturing data concerning limb movements of an experimental subject substantially as described herein and with reference to the Figures of the accompanying drawings.
41. A method of controlling the motion of a bipedal robotic device, comprising the steps of recording at preselected intervals data concerning angular displacement of the joints of the legs of a living subject and commanding each corresponding joint of the robotic device to adopt the same angular displacement at a corresponding time interval.
42. A motion control method as claimed in Claim 41, wherein the device is a robotic device as claimed in any one of Claims I to 16.
43. A motion control method as claimed in either Claim 41 or Claim 42, comprising the steps of capturing data as claimed in either Claim 39 or Claim 40.
44. A method for controlling the motion of a bipedal robotic device substantially as i described herein and with reference to the Figures of the accompanying drawings.