Artificial ankle-foot system with spring, variable-damping, and series-elastic actuator components
11273060 · 2022-03-15
Assignee
Inventors
- Hugh M. Herr (Somerville, MA)
- Kwok Wai Samuel Au (Mountain View, CA, US)
- Daniel Joseph Paluska (Silverthorne, CO, US)
- Peter Dilworth (Brighton, MA, US)
Cpc classification
A61F2002/5004
HUMAN NECESSITIES
A61F2002/763
HUMAN NECESSITIES
A61F2002/7635
HUMAN NECESSITIES
A61F2002/509
HUMAN NECESSITIES
A61F2002/503
HUMAN NECESSITIES
A61F2002/5075
HUMAN NECESSITIES
A61F2002/7645
HUMAN NECESSITIES
A61F2002/5073
HUMAN NECESSITIES
A61F2002/701
HUMAN NECESSITIES
A61F2002/6657
HUMAN NECESSITIES
International classification
B25J19/00
PERFORMING OPERATIONS; TRANSPORTING
B62D57/032
PERFORMING OPERATIONS; TRANSPORTING
Abstract
An artificial foot and ankle joint consists of a curved leaf spring foot member having a heel extremity and a toe extremity, and a flexible elastic ankle member that connects the foot member for rotation at the ankle joint. An actuator motor applies torque to the ankle joint to orient the foot when it is not in contact with the support surface and to store energy in a catapult spring that is released along with the energy stored in the leaf spring to propel the wearer forward. A ribbon clutch prevents the foot member from rotating in one direction beyond a predetermined limit position. A controllable damper is employed to lock the ankle joint or to absorb mechanical energy as needed. The controller and sensing mechanisms control both the actuator motor and the controllable damper at different times during the walking cycle for level walking, stair ascent, and stair descent.
Claims
1. A prosthetic, orthotic or exoskeletal ankle joint device comprising: a) a motor adapted to exert a torque about an ankle joint; b) a spring operatively coupled to the motor; c) an artificial sensory system comprising at least one gyroscope, and at least one accelerometer; and d) a processor linking the motor and the sensory system, wherein the processor computes, based on signals from the at least one gyroscope and the at least one accelerometer, an elevation of the device relative to an absolute position of a point at the ankle joint device, wherein the processor outputs a control sequence for stair descent in which during a swing phase an angle of the ankle joint is reoriented for toe-first contact upon detection of an elevation below zero relative to an initial position of the point at the ankle joint, and wherein the processor causes, subsequent to toe-first contact and during a stance phase of stair descent, a damping response to be applied to the ankle joint to thereby control ankle dorsiflexion movement.
2. The device of claim 1, wherein the artificial sensory system further includes a velocity sensor.
3. The device of claim 1, wherein the artificial sensory system further includes a position sensor that includes at least one sensor selected from the group consisting of a joint angular position sensor, motor shaft angular position sensor and an inertial absolute orientation position sensor.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
DETAILED DESCRIPTION OF THE INVENTION
(19) Two embodiments of an ankle-foot system contemplated by the present invention are described in detail below. The first embodiment (Embodiment 1) provides for elastic energy storage, variable-damping and a variable-orientation foot control. In addition to these capabilities, the second embodiment to be described includes a motor in series with a spring for providing joint spring stiffness control during the CP and CD phases, and a motive torque control during the PP phase of the walking cycle as described above.
Embodiment 1
(20) Mechanical Components
(21) The mechanical design of embodiment 1 is seen in
(22) The elastic leaf spring seen at 601 and 701 can be made from a lightweight, efficient spring material such as carbon composite, fiberglass or a material of similar properties. As seen in
(23) The variable-damper mechanism seen at 605 and 705 can be implemented using magnetorheological (MR), electrorheological (ER), dry magnetic particles, hydraulic, pneumatic, friction, or any similar strategy to control joint damping. For embodiment 1, a MR system is employed. Here MR fluid is used in the shear mode where a set of rotary plates shear thin layers of MR fluid. When a magnetic field is induced across the MR layers, iron particles suspended in carrier fluid form chains, increasing the shear viscosity and joint damping.
(24) The ribbon stop seen at 603 and 703 prevents the ankle joint from dorsiflexing beyond a certain maximum dorsiflexion limit, ranging from 0 to 30 degrees depending on ankle performance requirements. The ribbon stop is uni-directional, preventing dorsiflexion but not impeding plantarflexion movements.
(25) The actuator motor seen at 607 and 707 is a small, low-power electromagnetic motor that provides foot orientation control. The motor can exert a torque about the ankle joint (indicated at 711) to re-position the foot (the elastic leaf spring 601, 701) relative to the shank depicted at 713 when the foot is not in contact with the ground. As seen in
(26) Control System
(27) For a better understanding of the control sequence of the artificial ankle, a simplified 1D lumped parameter model of embodiment 1 seen in
(28) From
(29) Level-Ground Walking
(30) The control sequence of Embodiment 1 for level-ground walking is depicted in
(31) The state of each element of the ankle-foot system during the four phases of a level ground walking cycle are listed below:
(32) Controlled Plantar Flexion (
(33) Controlled Dorsiflexion (
(34) Powered Plantar Flexion (
(35) Swing Phase (
(36) The maximum dorsiflexion ankle torque during level-ground walking is in the range from 1.5 Ng to 2 Nm/kg, i.e. around 150 Nm for a 100 kg person {2}. With current technology, a variable-damper that can provide such high damping torque and additionally very low damping levels is difficult to build at a reasonable weight and size. Fortunately, the maximum controlled plantar flexion torque is small, typically in the range of 0.3 Nm/kg to 0.4 Ng. Because of these factors, a ribbon stop that engages at a small dorsiflexion angle such as 5 degrees would lower the peak torque requirements of the variable-damper since the peak controlled plantar flexion torque is considerably smaller than the peak dorsiflexion torque.
(37) During stair descent/downhill walking, the human ankle behaves like a damper from foot strike to 90° of dorsiflexion {11}. Beyond that, the ankle behaves like a non-linear spring, storing elastic energy during controlled dorsiflexion. Taking advantage of the biomechanics of the human ankle, it is reasonable to add a passive clutch for resisting dorsiflexion movements beyond 90°, thus allowing for a smaller sized variable damper. A ribbon stop is preferred as a unidirectional clutch because it is lightweight with considerable strength in tension.
(38) Stair Ascent
(39)
(40) The state of each element of the ankle-foot system during these three phases of a stair ascent are listed below:
(41) Controlled Dorsiflexion (
(42) Powered Plantar Flexion (
(43) Swing Phase (
(44) Stair Descent
(45) The control sequence for embodiment 1 for stair descent is depicted in
(46) The state of each element of the ankle-foot system during the four phases of stair descent are listed below:
(47) Controlled Dorsiflexion 1 (
(48) Controlled Dorsiflexion 2 (
(49) Powered Plantar Flexion (
(50) Swing Phase (
Sensing for Embodiment 1
(51) The ankle foot system preferably employs an inertial navigation system (INS) for the control of an active artificial ankle joint to achieve a more natural gait and improved comfort over the range of human walking and climbing activities.
(52) To achieve these advantages, an artificial ankle joint must be controlled to behave like a normal human ankle. For instance, during normal level ground walking, the heel strikes the ground first; but when descending stairs, it is the toe which first touches the ground. Walking up or down an incline, either the toe or the heel may strike the ground first, depending upon the steepness of the incline.
(53) A difficult aspect of the artificial ankle control problem is that the ankle joint angle must be established before the foot reaches the ground, so that the heel or toe will strike first, as appropriate to the activity. Reliable determination of which activity is underway while the foot is still in the air presents implacable difficulties for sensor systems presently employed on lower leg artificial devices.
(54) The present invention addresses this difficulty by attaching an inertial navigation system below the knee joint, either on the lower leg segment or on the artificial foot. This system is then used to determine the foot's change in elevation since it last left the ground. This change in elevation may be used to discriminate between level ground walking and descending stairs or steep inclines. The ankle joint angle may then be controlled during the foot's aerial phase to provide heel strike for level ground walking or toe strike upon detection of negative elevation, as would be encountered descending stairs or walking down a steep incline.
(55) Inertial navigation systems rely upon accelerometers and gyroscopes jointly attached to a rigid assembly to detect the assembly's motion and change of orientation. In accordance with the laws of mechanics, these changes may be integrated to measure changes of the system's position and orientation, relative to its initial position and orientation. In practice, however, it is found that errors of the accelerometers and gyros produce ever-increasing errors in the system's estimated position. Inertial navigation systems can address this problem in one of two ways: by the use of expensive, high precision accelerometers and gyroscopes, and by incorporating other, external sources of information about position and orientation, for instance GPS, to augment the purely inertial information. But using either of these alternatives would make the resulting system unattractive for an artificial ankle device.
(56) However, we have found that an unaugmented, purely inertial system based on available low cost accelerometers and rate gyros can provide sufficiently accurate trajectory information to support proper control of the angle of an actuated artificial ankle system.
(57) An Illustrative Control Algorithm
(58) Control of an actuated artificial ankle joint may be implemented as follows:
(59) A. During the foot flat (controlled dorsiflexion) phase of the walking cycle, reset and maintain the measured elevation to zero. When the foot is flat on the ground, its velocity and acceleration are zero. Thus, this particular foot posture serves as a reset point for the integration of angular and linear velocities in the estimation of absolute positions.
(60) B. During the push off phase, when powered plantarflexion begins, measure the upward and downward movements to determine the current elevation relative to the initial zero elevation during the flat foot phase;
(61) C. As long as the elevation remains above zero, maintain the foot orientation that will provide heelstrike; and
(62) D. If the elevation decreases below zero, reorient the angle ankle to provide toe-first contact.
(63) The foot flat phase may be detected by the absence of non-centrifugal, non-gravitational, linear acceleration along the length axis of the lower leg. Push off phase may be detected by the upward acceleration along the axis of the lower leg. Elevation>0 and elevation<0 phases are recognized from the change in relative elevation computed by the INS since the end of foot flat phase.
Embodiment 2
(64) Mechanical Design
(65) The mechanical design of Embodiment 2 is shown in
(66) The corresponding schematic of Embodiment 2 is seen in
(67) One of the main challenges in the design of an artificial ankle is to have a relatively low-mass actuation system that can provide a large instantaneous output power upwards of 200 Watts during Powered Plantar Flexion (PP) {2,11} Fortunately, the duration of PP is only 15% of the entire gait cycle, and the average power output of the human ankle during the stance phase is much lower than the instantaneous output power during PP. Hence, a catapult mechanism is a compelling solution to this problem.
(68) The catapult mechanism is mainly composed of three components: an actuator motor, a variable damper and/or clutch and an energy storage element. The actuator can be any type of motor system, including electric, shape memory alloy, hydraulic or pneumatic devices, and the series energy storage element can be any elastic element capable of storing elastic energy when compressed or stretched. The damper can be any type of device including hydraulic, magnetorheological, pneumatic, or electrorheological.
(69) With the parallel damper seen at 2305 in
(70) Control System
(71) The lumped parameter model of Embodiment 2 is shown in
(72) The control sequence of Embodiment 2 for level-ground walking will be discussed in the next section. Stair ascent/descent can be deduced from the earlier descriptions for embodiment 1, and thus, will not be described herein.
(73) Level-Ground Walking
(74) The control sequence of Embodiment 2 for level-ground walking is depicted in
(75) The state of each element of Embodiment 2 of the ankle foot system during the four phases of a level ground walking cycle are listed below:
(76) Controlled Plantar Flexion (
(77) Controlled Dorsiflexion (
(78) Powered Plantar Flexion (
(79) Swing Phase (
Sensing for Embodiment 2
(80) As with Embodiment 1, an inertial navigation system for the control of the active artificial ankle joint will be employed to achieve a more natural gait and improved comfort over the range of human walking and climbing activities. The manner in which these navigation sensors will be used is similar to that described for Embodiment 1.
(81) Sensing and Control
(82) As described above, investigations of the biomechanics of human limbs have revealed the functions performed by the ankle during normal walking over level ground, and when ascending or descending a slope or stairs. As discussed above, these functions may be performed in an artificial ankle joint using motors to act as torque actuators and to position the foot relative to the shin member during a specific times of walking cycle, using springs in combination with controllable dampers to act as linear springs and provide controllable damping at other times in the walking cycle. The timing of these different functions occurs during the walking cycle at times described in detail above. The specific mechanical structures, that is the combinations of motors, springs and controllable dampers used in these embodiments are specifically adapted to perform the functions needed, a variety of techniques may be employed to automatically control the motor and controllable dampers at the times needed to perform the functions illustrated, and any suitable control mechanism may be employed.
(83) The sensors used to enable general actuator operation and control can include:
(84) (1) Position sensors seen at 2902 in
(85) (2) A force sensor (strain gauges) to measure the actual torque borne by the joint as indicated at 2908.
(86) (3) Velocity sensors on each of the dampers (rotary encoders) as indicated at 2910 in order to get a true reading of damper velocity.
(87) (4) A displacement sensor on each spring (motor series spring and global damper spring) as indicated at 2912 in order to measure the amount of energy stored.
(88) (5) One or more Inertial Measurement Units (IMUs) seen at 2914 which can take the form of accelerometers positioned on skeletal members from which the processor 2900 can compute absolute orientations and displacements of the artificial joint system. For example, the IMU may sense the relative vertical movement of the foot member relative to its foot flat position during the walking cycle to control foot orientation as discussed above.
(89) (6) One or more control inputs manipulatable by a person, such a wearer of a prosthetic joint or the operator of a robotic system, to control such things as walking speed, terrain changes, etc.
(90) The processor 2900 preferably comprises a microprocessor which is carried on the ankle-foot system and typically operated from the same battery power source 2920 used to power the motor 2930 and the controllable dampers 2932 and 2934. A non-volatile program memory 2941 stores the executable programs that control the processing of the data from the sensors and input controls to produce the timed control signals which govern the operation of the actuator motor and the dampers. An additional data memory seen at 2942 may be used to supplement the available random access memory in the microprocessor 2900.
(91) Instead of directly measuring the deflection of the motor series springs as noted at (4) above, sensory information from the position sensors (1) can be employed. By subtracting the ankle joint angle from the motor output shaft angle, it is possible to calculate the amount of energy stored in the motor series spring. Also, the motor series spring displacement sensor can be used to measure the torque borne by the joint because joint torque can be calculated from the motor series output force.
(92) Many variations exist in the particular sensing methodologies employed in the measurement of the listed parameters. Although this specification describes preferred sensing methods, each has the goal of determining the energy state of the spring elements, the velocities of interior points, and the absolute movement pattern of the ankle joint itself.
REFERENCES
(93) The following published materials provide background information relating to the invention. Individual items are cited above by using the reference numerals which appear below and in the citations in curly brackets. {1} Palmer, Michael. Sagittal Plane Characterization of Normal Human Ankle Function across a Range of Walking Gait Speeds. Massachusetts Institute of Technology Master's Thesis, 2002. {2} Gates Deanna H., Characterizing ankle function during stair ascent, descent, and level walking for ankle prosthesis and orthosis design. Master's thesis, Boston University, 2004. {3} Hansen, A., Childress, D. Miff, S. Gard, S. and Mesplay, K., The human ankle during walking: implication for the design of biomimetric ankle prosthesis, Journal of Biomechanics (In Press). {4} Koganezawa, K. and Kato, I., Control aspects of artifical leg, IFAC Control Aspects of Biomedical Engineering, 1987, pp. 71-85. {5} Herr H, Wilkenfeld A. User-Adaptive Control of a Magnetorheological Prosthetic Knee. Industrial Robot: An International Journal 2003; 30: 42-55. {6} Seymour Ron, Prosthetics and Orthotics: Lower limb and Spinal, Lippincott Williams & Wilkins, 2002. {7} G. A. Pratt and M. M. Williamson, “Series Elastic Actuators,” presented at 1995 IEEE/RSJ International Conference on Intelligent Robots and Systems, Pittsburgh, Pa., {8} Inman V T, Ralston H J, Todd F. Human walking. Baltimore: Williams and Wilkins; 1981. {9} Hof. A. L. Geelen B. A., and Berg, Jw. Van den, “Calf muscle moment, work and efficiency in level walking; role of series elasticity,” Journal of Biomechanics, Vol 16, No. 7, pp. 523-537, 1983. {10} Gregoire, L., and et al, Role of mono- and bi-articular muscles in explosive movements, International Journal of Sports Medicine 5, 614-630. {11} Koganezawa, K. and Kato, I., Control aspects of artifical leg, IFAC Control Aspects of Biomedical Engineering, 1987, pp. 71-85. {12} U.S. Pat. No. 6,517,503 issued Feb. 11, 2003.
CONCLUSION
(94) It is to be understood that the methods and apparatus which have been described above are merely illustrative applications of the principles of the invention. Numerous modifications may be made by those skilled in the art without departing from the true spirit and scope of the invention.