Lower limb exoskeletons are developed to increase human strength or the help people in their revalidation process. LEX has been developed over the last years as a research platform to experiment with ways to increase the efficiency of the gait cycle. In this research a computer model of walking person in an LEX was developed using recorded motion capture data and de modeling software ADAMS and LifeMod. This model was compared to other gait cycle studies and it was concluded that the model is good enough to use in further research, although it has some shortcomings and is not accurate enough to prove new theories. I was also shown that the increase of energy of one gait cycle while walking in an un-actuated exoskeleton is 29%.
 |
Figure 1 – LEX, the 6-DOF lower limb exoskeleton developed by UCSC, Bionics Lab. |
 |
Figure 4 – Motion Capture Software, PhaseSpace system. |
The software package MD Adams R3 with the add-on package LifeMod was used to develop the model and do the simulations. Adams is a dynamic simulation package in which the exoskeleton was imported from SolidWorks drawings. All the joints had to be properly defined and the intertia specified. LifeMod has a big database with human models which can be used in ADAMS. Since LEX was designed for a 1.8m man, a human model of 1.8m, 77kg and 300 months old was used from the GeBOD database.
To create a walking person the body segments were automatically created by LifeMod according to the above mentioned specifications. A base set of joints available in LifeMod was used as well. Motion agents were attached to the model which correspond to the measured data markers as can be seen in Figure 3a.
Out of one data recording of 15 seconds walking, a representative gait cycle has been chosen. Figure 7, Figure 8 and Figure 9 show the characteristics for the ankle, knee and hip during a gait cycle expressed in the angle, torque and power.. The x-axis shows the percentage of the gait cycle.
The joint angles are defined as shown in Figure 6.
 |
Figure 6 – Definition of the joint angles |
The average power over one gait cycle has been calculated using 9 gait cycles. The average energy for one gait cycle without an exoskeleton was 168J and with an un-actuated exoskeleton 216J. That is an increase of 29%. The energy consumption increased for all the joints: ankle +63%, knee +34%, hip +16%. The portions of negative energy did also increase, in total with 32%.
 |
Figure 7 – Gait cycle profiles for the ankle without an exoskeleton and with an un-actuated exoskeleton |
 |
Figure 8 – Gait cycle profiles for the Knee without an exoskeleton and with an un-actuated exoskeleton. |
Table 1 – Ankle data compared to other studies. Shows between what values the data is or what the extreme of a single peak is.
| LEX | Other data | |
| Angle (deg) | -22 to -10 | between -20 to 14 |
| Torque (Nm) | -122 (peak) | around -120 (peak) |
| Power (W) | 105 (peak) | peak between 150 to 250 |
Table 2 – Knee data compared to other studies. Shows between what values the data is or what the extreme of a single peak is.
| LEX | Other data | |
| Angle (deg) | -52 | peak between 55-70 |
| Torque (Nm) | -80 to 80 | between -40 to 60 |
| Power (W) | -240 to 80 | between -150 to 50 |
 |
Figure 9 – Gait cycle profiles for the Hip without an exoskeleton and with an un-actuated exoskeleton. |
Table 3 – Hip data compared to other studies. Shows between what values the data is or what the extreme of a single peak is.
| LEX | Other data | |
| Angle (deg) | -20 to 18 | between -22 to 30 |
| Torque (Nm) | -150 to 120 | between -70 to 60 |
| Power (W) | -65 to 390 | between -60 to 110 |
Taking all the foregoing into consideration it can be concluded that a workable model has been created that can be used for further research. However, it’s very important to be aware of the shortcomings of this model. The model is reliable enough to use in experiments for implementing the pneumatic actuation, to do optimization studies or energy analysis. It is probably not possible to prove theories with this model, but it can help to point in a certain direction or be used to conform a line of thought. Implementation of the found parameters in the real setup is needed to prove those theories. Too many parameters which can arbitrarily be adjusted and uncertainties during the measures make it difficult to draw very hard conclusions from this model. This all makes the model maybe too complicated for its purpose and too sensitive for a change in parameters.
Furthermore this research shows that walking with the un-actuated exoskeleton LEX increases the energy consumption with 29% and influences the walking pattern.
