The SEA provides force control by controlling the extent to which the series spring is compressed. A linear potentiometer measure the deflection of the spring. The advantages of the SEA are that it has low impedance, the motor is isolated from shock loads, and the effects of backlash, torque ripple, and friction are filtered by the spring. A further advantage is that the SEA exhibits stable behavior while in contact with most environments, even when in parallel with a human limb.

It is shown that during controlled plantar flexion, normal ankle has a constant linear stiffness and can be modeled as a linear rotational spring where ankle moment is proportional to ankle position. Thus, during this phase of walking (Contact1) it is desired to control a linear stiffness at the orthotic ankle joint. Since the stiffness of the ankle is depends on the walking velocity, it is essential to measure the ground reaction force generated at the moment of forefoot impact after each walking step. For this purpose Ultraflex system us used where a series of capacitive force transducers are placed on the bottom of the AFO (two sensors beneath the heel and four beneath the forefoot region).

To estimate the state of walking, a single foot switch is place in the heel of a shoe worn with the orthosis to detect heel strike approximately 30 ms earlier that the Ultraflex force sensors. There is a rotary potentiometer used to determine the angle between the shank and the foot.

Gait speed is an important step-to-step gait variation which affects the stiffness of the ankle. Therefore the controller needs to measure this parameter to be able to respond and adapt properly. There are different criterions to estimate the walking speed which the time of foot-contact (time between heel strike and toe off) is the most trustable one.

The result of this AAFO tested on a subject at MIT is shown in this video clips. (Click on the picture to play- You may need RealPlayer).