Text

Introduction: Exoskeletons (EXOs) for supporting physical work reduce acute physical strain in their respective target regions. In contrast, little is known about the side effects of EXOs, such as the effects of load redistribution to other (non-targeted) areas of the musculoskeletal system and the associated health risks. We therefore provide two examples (Study A and B) of side effects caused by two different EXOs.

Methods:

Study A: 29 subjects performed a manual task in a static forward bent upper body posture (40° inclination), with and without a back support exoskeleton (EXO-Back). Ground reaction forces, EXO-Back support torque, and the position of the femur, tibia, and spine were measured continuously and incorporated with the subjects’ body dimensions into a specially developed biomechanical model that allows continuous determination of knee joint loading.

Study B: 45 male subjects with and without an exoskeleton supporting the lower limbs (EXO-LowerLimbs) lifted a 3 kg object at the lateral end of their reach range and placed it at different angles (120°, 150°, and 180°) in the transverse plane. Postural stability, as the shortest distance between the body’s centre of gravity (centre of pressure) and the nearest edge of the support surface, was measured. A distance <= 0 indicates an immediate fall. In addition, the stance stability of 8 male subjects with and without EXO-LowerLimb was determined while a force was applied posteriorly in the sagittal plane (external disturbance moment). The torque at which the subjects fell backwards was evaluated.

Results: Using the model developed in Study A, the vertical and horizontal forces acting on the hip, knee, and ankle joints could be calculated as a function of body proportions and the support torque of the EXO-Back. Using the EXO-Back led to a median force of 800 to 900 N acting on each knee joint. The horizontal component, posterior forces on the knees with the EXO-Back were up to 2 N in the median and 25 N in the upper quartile. With the EXO-LowerLimb, the postural stability was significantly lower, but still approximately 17 mm for each position. When external perturbing forces were applied, the fall inducing moment decreased from 50 Nm without to 30 Nm with the EXO-LowerLimb.

Discussion: The main side effect of the EXO-Back used with respect to the knee joint load is limited to a shift of the horizontal force with a shearing effect in the posterior direction. The forces occurring in the study are classified ad hoc as “probably not harmful” but are linearly related to the relatively low support moment of the exoskeleton studied but could be much higher depending on the EXO support.

With an EXO-LowerLimb, a person’s compensation strategies to counteract falling in the event of a loss of balance are compromised. Lateral stability with EXO-LowerLimb is still considered adequate under the simulated working conditions. However, external forces directed to the rear, such as those that may be generated by collaborative robots, have the potential to cause the user to fall backwards.

Conclusion: The examples presented should encourage people not to overestimate the possible side effects of exoskeletons, but to consider them in evaluation studies.