0
Research Papers

An Active Foot-Ankle Prosthesis With Biomechanical Energy Regeneration

[+] Author and Article Information
Joseph K. Hitt

Department of Civil and Mechanical Engineering, West Point, New York, NY 10996Joseph.Hitt@usma.edu

Thomas G. Sugar

Mechanical and Aerospace Engineering, P. O. Box 876106, Tempe, AZ 85283thomas.sugar@asu.edu

Matthew Holgate

Mechanical and Aerospace Engineering, P. O. Box 876106, Tempe, AZ 85283matthew.holgate@asu.edu

Ryan Bellman

Mechanical and Aerospace Engineering, P. O. Box 876106, Tempe, AZ 85283Ryan.Bellman@asu.edu

J. Med. Devices 4(1), 011003 (Mar 26, 2010) (9 pages) doi:10.1115/1.4001139 History: Received August 19, 2009; Revised January 22, 2010; Published March 26, 2010; Online March 26, 2010

A unique, robust, robotic transtibial prosthesis with regenerative kinetics was successfully built and a 6-month human subject trial was conducted on one male below-the-knee amputee under linear walking conditions. This paper presents the quasistatic system modeling, DC motor and transmission modeling and analyses, design methodology, and model verification. It also outlines an approach to the design and development of a robotic transtibial prosthesis. The test data will show that the true power and energy requirement predicted in the modeling and analyses is in good agreement with the measured data, verifying that the approach satisfactorily captures the physical system. The modeling and analyses in this paper describes a process to determine an optimal combination of motors, springs, gearboxes, and rotary to linear transmissions to significantly minimize the power and energy consumption. This kinetic minimization allows the downsizing of the actuation system and the battery required for daily use to a self-portable level.

FIGURES IN THIS ARTICLE
<>
Copyright © 2010 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Desired spring deflection is achieved by controlling the motor position and capitalizing on the cyclical nature of gait. As the tibia rotates over the stance foot, springs are extended. Simultaneously, the motor extends the springs to achieve the desired spring deflection and the forces required to generate the required ankle moment for walking. This inverted pendulum with a lumped mass illustrates the regeneration energy with use of a spring in series with a motor. Computer aided design model of the prototype is illustrated on the right.

Grahic Jump Location
Figure 2

A two degrees-of-freedom model with a seismic excitation representing the motor excitation, a torsion spring for the keel, and a helical (linear) spring between the lever and the motor is shown. The moment due to the keel is a function of φ(t) and the moment due to the spring is a function of Θ(t)l−x(t). The moment at the ankle is from published information determined using inverse dynamics of motion capture and force plate test data as published in Ref. 17.

Grahic Jump Location
Figure 3

This diagram illustrates the flow of power and energy from the battery to the user. Significant amount of energy is lost due to inefficiency in the mechanisms, motor, inertia, friction, etc. Proper selection and design can drastically improve overall system efficiency. Note that the system efficiency is defined as average output power to the user/average input power from the battery. The inertia and friction box is assumed to be linked to all of the mechanical elements.

Grahic Jump Location
Figure 4

This diagram illustrates the flow of energy from the battery to the user for the robotic tendon model. Even though significant amount of energy is lost due to inefficiency in the mechanisms, motor, inertia, friction, etc., the spring and the regenerative energy that it harnesses is nearly 100% efficient and accounts for the main share of the output energy. This method also allows for a smaller motor, battery, and transmission system.

Grahic Jump Location
Figure 5

3D plot of the RE40 motor efficiency as a function of motor torque (Nm) and motor angular velocity (rpm) (18). Notice that the highest efficiency of 91% is only achieved at a narrow range of torque and angular velocity. Operating the RE40 at speeds lower than 2000 rpm or torque above 0.2 Nm will significantly degrade the motor efficiency. Illustrated in the figure are two points on the mesh.

Grahic Jump Location
Figure 6

The three motor efficiency 3D plots overlapped. The blue is the RE 40, the red is the RE 30, and the green is the EC 30. The red surface ends at 1.02 Nm on the horizontal axis describing torque. The vertical axis measures efficiency.

Grahic Jump Location
Figure 7

Shown for one gait cycle are the electric power Pe, blue, power at the nut Pm, black, the output power Po, red, and the power consumed due to rotor and gear inertia Pj. The area in the blue circle highlights the inertia effect and its power curve. The area in the yellow circle highlights the efficiency effect and the larger electric power curve.

Grahic Jump Location
Figure 8

Energy surface plots for the RE 40 motor in combination with different lead screw pitch and gear ratio. The best result, 53.1 J/step, is at 4.6 GR and 3 rev/in. lead screw.

Grahic Jump Location
Figure 9

Left figure is the isometric and side views of current design as modeled in SOLIDWORKS . The RE40 motor coupled with the robotic tendon provides a dynamic moment about the ankle joint. The right figure is a photograph of SPARKy on a male transtibial amputee test subject.

Grahic Jump Location
Figure 10

Electric power, determined from voltage and current input readings, for the test subject on a treadmill at 1 m/s with 9 cm lever arm and 36 KN/m spring stiffness. Note that there is variability in each step, even though the subject is on a treadmill that is set at 1 m/s (2.2 mph).

Grahic Jump Location
Figure 11

The same electric power data set as in Fig. 1. The dotted line, left figure, is the predicted Pe from Eq. 10, and the solid lines are the raw measurements. The right figure shows the mean and standard deviation of the Pe data and the model, as annotated.

Grahic Jump Location
Figure 12

Measured power out Po and power at the nut Pm with 36 KN/m spring, 9 cm lever at 1 m/s (2.2 mph). The left figure shows the raw data for the Po and the Pm for multiple gait cycles. The right figure shows the mean and standard deviation of the data and its corresponding models, as annotated. Note that the device achieves a very high level of power amplification of 3.7. This is the unique advantage of a robotic tendon.

Grahic Jump Location
Figure 13

Measured ankle joint angle for the same test series: 9 cm lever arm, 36 KN/m spring, and 1 m/s (2.2 mph). The left figure shows the raw data for the ankle joint angle for multiple gait cycles and the model. The right figure shows the mean and standard deviation of the data and the model, as annotated. Note that the device provides ankle joint motion that is comparable to able-bodied gait.

Grahic Jump Location
Figure 14

Measured ankle joint moment: 9 cm lever arm, 36 KN/m spring, and 1 m/s (2.2 mph). The left figure shows the raw data for the ankle joint moment for multiple gait cycles and the model. The right figure shows the mean and standard deviation of the data and the model, as annotated. Note that the device provides ankle joint moment that is comparable to able-bodied gait.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In