Wire-driven Running Reinforcement Shoes

1. Introduction

1.1 Background and Objectives of the Study

Humans have the advantage of bipedal locomotion, which allows them to have free use of their hands. However, when it comes to running, there is a physical constraint that limits the use of only the leg muscles. Considering that even cheetahs, known for their fast running speed, utilize all four limbs, it can be hypothesized that if humans could utilize the muscles of other body parts in addition to the legs while running, they could achieve faster running speeds[1].

Based on this idea, we aimed to utilize the muscular strength of the arms, which play a role in maintaining balance and facilitating smooth leg movements during running. The objective of this study was to develop an auxiliary device that attaches to the shoes and transfers the muscular strength of the arms to the soles, generating additional propulsive forces to increase running speed.

1.2 Significance of the Study

While there are devices like roller skates that can increase human running speed, such wheeled devices cannot be used on uneven surfaces like dirt roads. However, the device developed in this study can maintain the typical running motion even on uneven surfaces and effectively increase running speed.

Furthermore, the device assists running without requiring additional external power and utilizes muscles from body parts that were not previously involved in running. This highlights the significance of the device as an enhancement tool that relies solely on human power to improve running performance.

2. Methodology

2.1 Device Design

The designed device, named Wire-driven Running Reinforcement Shoes (WRRS), utilizes wires as power transmission elements to enhance running performance.

The components of the WRRS, which are attached to the sides of the shoes, can be divided into three main parts. These include the wire and pulley mechanism for transferring arm power to the ankles, the internal gear mechanism that amplifies the output torque of the wire, and the pad section on the shoe sole that generates propulsive forces by pushing off the ground with the amplified rotational output.

2.1.1 Mechanism

To maintain balance in the body, the hands and feet move in opposite directions while running. As the feet push off the ground, the wrists naturally move away from the ankles, pulling the connected wire. In this process, a mechanism has been devised to transmit rotational output to the shoe sole through a gear box located in the shoe, thereby obtaining propulsion force.

2.1.2 Design Constraints and Objectives

When setting the wire pulling as the input and the rotation of the shoe pad as the output, increasing the gear ratio of the gear box is crucial for achieving high propulsion force. To determine the specific requirements for the gear output ratio, the necessary rotation angle of the shoe pad and the constraints on the wire length that can be pulled during running motion need to be established.

The moment when the wire is most pulled corresponds to the moment when the distance between the ankle and wrist is the greatest, which is similar to the toe-off moment. Additionally, this distance is minimized during foot-strike[2]. Therefore, the length difference between foot-strike and toe-off was measured. Based on measurements taken during running motion, the length of the pulled wire ranged from 18cm to 20cm, and this range was used to define the operational range.

The flexion angle of the shoe is known to have an impact on propulsion force, with the highest propulsion force observed at a moderate flexion angle of 50 degrees[3]. Considering this, the output rotation angle of the pad was set between 20 degrees and 30 degrees. Excessive rotation angle can lead to instability in posture at the flexion point of the foot.

2.1.3 Gear box : Planetary Gear Train

The internal gear box of the main body was designed by modifying the structure of a Planetary Gear Train[4].

In order to maintain high gear output for supporting body weight and pushing against the ground, while also ensuring compatibility with shoe attachment, the diameter of the gear box needs to be small. To achieve this, a Double Layer Planetary Gear system was designed, where the Planetary Gear is arranged in two layers, resulting in a decreased diameter for the same gear ratio.

Comparing to the gear ratio of 5.8:1 used in the Cheetah Robot’s actuators developed at MIT[5], assuming the same diameter of the disk, the Double Layer Planetary Gear structure achieves a gear ratio output of 1:10.5 with the same number of gear engagements, allowing for approximately 1.8 times higher torque compensation.

Figures 2 and 3 depict cross-sectional views of the Double Layer Planetary Gear. It shows the difference from conventional gear systems as the Planetary Gear is divided into two layers, providing rotational output through the Ring gear.

*[Fig 2] Cross-sectional view of the Double-layered Planetary gear_1*

*[Fig 3] Cross-sectional view of the Double-layered Planetary gear_*2

Considering the high precision and stress concentration of gears, it was determined that 3D printing has limitations for producing them. Therefore, a spur gear was chosen, taking into account the bending and torsional loads analyzed. The gear was selected based on the Standard Gear with a module of 1.0mm, considering the compatibility with 3D printed parts and choosing the specifications with keyway and hub.

2.1.4 Wire Output Section

To operate the gear mechanism with the wire, a pulley component was designed. A rubber band was utilized to reset the pulley to its reset position when transitioning from the toe-off motion to the foot-strike motion. To prevent malfunction caused by the wire and rubber band overlapping, the pulley was positioned on the opposite side of the pulley shaft to allow simultaneous winding and unwinding. As a result, the likelihood of the wire and rubber band getting tangled was reduced, and the generation of moments was brought closer to the bearing for stable rotation.

2.2 Analysis

2.2.1 Simulation

To analyze the gear’s motion more accurately, the specific rotation conditions of the pulley were determined, and the operation was simulated using the Simulation program within Solidworks. Based on the calculated angular position of the pulley due to wire tension, angular velocity and acceleration were set using a 3rd order polynomial for the simulation.

2.2.2 Ansys Analysis

First, the entire model was created using the SOLIDWORKS program, and the gear part of the overall model was analyzed using the ANSYS analysis program. The maximum allowable moment and deformation at that point were compared for two different materials: PLASTIC (ABS) and SUS 304. If the applied external force is smaller than the yield strength of the material, elastic deformation occurs, and the material returns to its original shape. However, if the applied external force is stronger than the yield strength, the material undergoes plastic deformation and does not return to its original shape. Therefore, the maximum allowable stress for each material was calculated by dividing the yield stress by a safety factor of 1.25 (maximum stress = yield stress / safety factor). The maximum allowable moment, which occurs when the maximum stress is reached, was also calculated.

2.2.2.1 Plastic Material

Yield Stress: 2.744×10⁷Pa
Safety Factor: 1.25
Maximum Allowable Stress = Yield Stress / Safety Factor: 2.744×10⁷ / 1.25 = 2.1952×10⁷Pa
Maximum Allowable Moment = 1.35 Nm

2.2.2.2 SUS 304 Material

Yield Stress: 2.5×10⁹Pa
Safety Factor: 1.25
Maximum Allowable Stress = Yield Stress / Safety Factor: 2.5×10⁹ / 1.25 = 2.0×10⁹Pa
Maximum Allowable Moment = 120 Nm

	Plastic (ABS)	SUS 304
Yield Stress (Pa)	2.744×10⁷	2.5×10⁹
Safety Factor	1.25	1.25
Maximum Allowable Stress (Pa)	2.1952×10⁷	2.0×10⁹
Maximum Allowable Moment (Nm)	1.35	120
Deformation (m)	5.89×10^-5	5.50×10^-7

[Table 1] Yield Stress, Maximum Stress, Maximum Moment, and Total Deformation

2.3 Manufacturing

2.3.1 3D Printing

To fabricate the model, the Mark2 3D printer model from Markforged was utilized for 3D printing.

*[Fig 6] 3D Printing with Markforged ‘Mark 2’*

3. Conclusion

In order to enhance human running ability by utilizing other body parts that were not previously associated with running, we have designed and fabricated a device that operates smoothly.

Although the goal of the project has been achieved in the design and fabrication phase, further research can be conducted to quantitatively evaluate the improvement in running ability using this device.

Furthermore, to reset the pulley to the initial position when transitioning from the Toe-off motion to the Foot-strike motion, a rubber band was used. However, there is a possibility that the elastic modulus of the rubber band may change over time due to prolonged device use. Therefore, replacing the rubber band with a torsional spring that has an appropriate elastic modulus can improve the device’s lifespan.

Considering the high precision and stress concentration of gears, 3D printing has limitations. Therefore, we have selected spur gears considering the axial and torsional loads by analyzing the axial and torsional loads and selecting the gear specifications based on keyway hole compatibility.

Plastic (ABS) was used as the material for the model, and the maximum stress, maximum moment, and deformation were found to be 2.1952×10⁷ Pa, 1.35 Nm, and 5.89×10^-5 m, respectively. When SUS 304 was used as the material, the values were 2.0×10⁹ Pa, 120 Nm, and 5.50×10^-7 m, respectively. Based on this, it can be inferred that using Plastic as the material, which can be easily fabricated using 3D printing, would result in lower propulsive force compared to using SUS 304. However, the model we created, which used off-the-shelf gears (SUS 304), can achieve strong propulsive force. Nevertheless, it is heavier compared to when using Plastic, which may increase the supporting force required. If the model is made with a lighter and higher yield strength material such as Carbon Plate, more ideal results can be expected.

Our final fabricated model had a flat bottom surface for the pad. However, a new design method was proposed by Lee et al. [9], which demonstrated that modifying the design factors of the pattern (width of peaks, width of valleys, height of peaks, slope of peaks, slope of pattern, etc.) can affect the friction coefficient. It was observed that increasing the width of the valleys resulted in a decrease in the friction coefficient. However, the other factors varied differently depending on the pattern, indicating that the influence of the uncertainties in the specimen fabrication process had a greater impact on the friction coefficient than the pattern itself. The overall average friction coefficient obtained from the four basic patterns showed that the average friction coefficient of the O-pattern was the highest among the straight-line pattern, W-pattern, O-pattern, and wave pattern[9]. Therefore, by adjusting the width of the valleys of the O-pattern, which has the highest average friction coefficient, and incorporating it into the bottom surface of the model using the new design method [8], it is expected to increase the frictional force with the ground and obtain stronger propulsive force with the same force.

Lastly, the operability of the fabricated device was evaluated. The obtained results were in accordance with the target ratio of output to input set for the gear box during the initial design stage. When the wire was pulled 19 cm during the Toe-off motion, the pad rotated by 17 degrees. Additionally, during the Foot-strike motion, it was confirmed that the pulley returned to the initial position smoothly due to the rubber band.

4. Reference

Bramble, D. M., & Lieberman, D. E. (2004). Endurance running and the evolution of Homo. Nature, 432, 345–352.
Simpson, C. S., Welker, C. G., Uhlrich, S. D., Sketch, S. M., Jackson, R. W., Delp, S. L., Collins, S. H., Selinger, J. C., Hawkes, E. W. (2019). Connecting the legs with a spring improves human running economy. Journal of Experimental Biology, 222(17).
Mok, S., Kwak, C., & Kwon, O. (2004). The effect of midsole hardness of running shoes on foot flexion angle. Korean Journal of Sport Biomechanics, 14(2), 85-103.
Lévai, Z. (1968). Structure and analysis of planetary gear trains. Journal of Mechanisms, 3(3), 131–148.
Seok, S., Wang, A., Chuah, M. Y., Otten, D., Lang, J., & Kim, S. (2013). Design Principles for Highly Efficient Quadrupeds and Implementation on the MIT Cheetah Robot. ICRA.
Lee, J. H. (2009). Comparative analysis of the kinetic properties between functional spring shoes and regular athletic shoes. Korean Journal of Sport Biomechanics, 19(3), 581-592.
Lee, J. H., Sung, B. J., & Song, J. H. (2007). Comparative analysis of the kinematics between functional spring shoes and regular athletic shoes. Korean Journal of Sport Biomechanics, 17(1), 99-109.
Kim, T. H. (2008). The effects of shoe size on plantar pressure, ground reaction force, and muscle fatigue during walking and running. Journal of the Korean Society of Physical Therapy, 15(1), 1-11.
Lee, J. N. (2008). A study on the relationship between the pattern shape of shoe sole and the friction coefficient. Korean Journal of Sport Biomechanics, 18(3), 1-10.