Personal tools

Interested in disability history? Check out what happened Today in AT History!

Robotic Ambulation Devices

From ATWiki

Jump to: navigation, search


Robotic Ambulation Devices


In the late 1990’s a new type of rehabilitation techniques were developed through the support of the Department of Veterans Affairs and the National Institute of Health. These new techniques called Constraint-Induced Movement Therapy (CI Therapy) involved isolating and separating the movements of the arms. For the less affected arm it was immobilized 90% of the time using a sling. Meanwhile the more affected arm was placed in a training regime using concentrated repetitive tasking (Taub, Uswatte, and Pidikiti, 1999).

After performing initial studies, research suggested that CI therapy produces task specific cortical reorganization. Research has demonstrated that CI Therapy leads to a recruitment of neurons in the process of moving to the opposite side of the brain with regard to the affected arm. This reorganization may serve as a neurological basis for permanent enhance in ability level of the arm. After the initial success using the upper arm, body weight supported harnesses were developed for the use of CI therapy on the lower limbs. It is from these harnesses that the robotic mechanisms were developed using the treadmill as a methodology for promoting locomotion in rehabilitation patients (Taub, Uswatte, and Pidikiti, 1999).

Lokomat DGO

Hocoma’s Lokomat Driven Gait Orthosis, one of the most commonly used robotic devices, consists of mechanisms which are used to help positioning of specific limbs. The Lokomat DGO synchronizes the speed of the robotics with the movement of the treadmill. The motors guiding the hips and knees are preprogrammed to flex and straighten according to a normal locomotive pattern with the normal physics of a gait pattern. To get an idea of the neurological impairment level of each patient American Spinal Cord Injury Association (ASIA) standards are used as references. According to the standard, muscle testing of five key groups is necessary. These groups are hip flexors, knee extensors, ankle dorsiflexors, great toe extensors, and ankle plantar flexors. These values are then added together to come up with an overall lower extremity score (Winchester et al., 2005).

For individuals that have difficulty standing or are unable to stand, after the individual is attached to the harness they are brought to standing using a “motor driven winch” (Winchester et al., 2005). Under normal circumstances it can take 2-3 therapists to correctly configure the patient, and even then there is the issue of temporality. It becomes extremely challenging as well as taxing on the therapists and the patients to maintain the correct positioning and move the person at the same time. This can assist in helping patients experience the senses of the different parts of their body and how they are working together, known as proprioceptive input (Winchester et al., 2005).

These devices can help not only in instruction scenarios on proper gait training, but also can use precise monitoring to help monitor patient progress quantitatively. The Lokomat system has a display that is a biofeedback system used during actual rehabilitation time. Force measuring devices are consistently assessing the workload between the patient and the machine (Winchester et al., 2005). These devices can also assist in the patient psychological well being in that they can help to promote motivation and provides means for rehabilitation both inside the therapy gym as well as within the home environment. These devices can provide specific scenarios that can be dynamic in the working relationship between the patient and the technology. They can serve as invention tools for the recovery of functionality. The ability to gain even some locomotive control capability after a significant SCI injury can help to continue developing how the sensory system learns and responds to stimulus (Edgerton & Roy, 2009).

Rehabilitation with Robotic Devices

In order for rehabilitation to occur via a robotic device, the spinal circuitry must be active. The problem with this is that to keep the spinal circuitry active, the experimenter cannot allow for consistent patterns, but rather needs to keep the circuitry guessing. The reason for this is that spinal circuitry tends to be less responsive to repeated stimuli. This makes sense because as a stimulus continues to occur the body adapts to it. If a consistent pattern is repeated the sensory recognition of that pattern decreases dramatically in a 2-3 minute span (Edgerton & Roy, 2009).

One of the ways to counteract this pattern recognition and forcing the spinal circuitry to stay engaged is to provide a system where the level of assistance varies according to the needs of the individual patient. This is similar to traditional physical therapy rehabilitation where a physical therapist provides just enough assistance to enable proper locomotive movement. The advantage of robotic devices is the ability to consistently change the level of assistance causing more or less weight bearing and muscle coordination to be control by the sensory system. The robotic device can also focus on individual muscle recruitment processes of the locomotive action to encourage proper response, such as the swing phase of a step (Edgerton & Roy, 2009).

During the process of reacquiring the ability to step, further optimization on the amount of variation between the stepping movements. Edgerton and Roy (2009) hypothesize that how much variation that will be needed is a function of the injury and recovery progress at the time. At the beginning the variation between steps could be quite large in an effort to gain familiarity to the gross coordination. Then as the individual begins to acquire the skill of walking, the amount of variability can be reduced until the variability would match that of everyday pre-injury locomotion. In addition the ability level of the individual serves as a consideration when determining the magnitude of variation which goes for non SCI patients as well as those with injuries (Edgerton & Roy, 2009).

Robotic Ambulation User Study

The systematic control of robotic devices allows for a greater specificity of therapy within rehabilitation. The more components of the locomotive system that can be controlled, the greater the potential of the synergistic action and products to accommodate this are currently under development (Edgerton & Roy, 2009). Winchester et al. (2005) find promising results in four patients following 12 weeks of robotic treadmill therapy. Using toe flexion and ankle plantar flexion as their baseline they are able to suggest empirically that task specific training may result in improvement in the supraspinal network for incomplete SCI. Regardless of the individual condition all patients show an increase of activity within the motor cortices following therapy. Specifically the patients have an increase in the foot and toe areas respectively (Winchester et al., 2005).

An interesting result to this study is the correlation between the functional gains as a result of therapy, and the magnitude of changes within the cerebellum with regard to activation. During the process of walking, the cerebellum coordinates the feedback from the periphery sensory receivers as well as receiving process specific commands for the motor aspect of stepping. It then proceeds to consistently adjust the walking patterns based on the input it receives from the senses. This activation appears to have a direct effect on the patient’s ability to walk over ground (Winchester et al., 2005).

The two individuals who had the greatest activation also achieved independent ambulation with assistive devices. On the other end of the spectrum, the patient with the lowest activation level was unable to obtain the ability to walk over ground. It is a possibility that these activations could be a product of motor learning or improved performance from the task at hand. However due to lack of differentiation of performance before and after the task, attribution toward motor learning appears more likely (Winchester et al., 2005).

Furthermore a connection is evident between the time gap between injury and time of rehabilitation. The individuals who have the shortest time since injury also demonstrate the largest gain with regard to locomotive ability. According to Winchester et al. as time passes patients “…may lack the plasticity necessary to develop and use afferent and deferent pathways necessary to benefit from this rehabilitation”. It is also important to note that the subjects who came into the study closest from the time of injury, could have achieved higher lower extremity motor scores due to residual input left from prior to their injury that someone who has been injured longer may not have (Winchester et al., 2005).

Honda’s Locomotive Devices

Development of locomotive assistive devices has moved into the private sector as the need for individuals needing the ability to walk becomes greater. Honda innovation, which has already created an advance humanoid robot, ASIMO, has applied its unique cooperative control technology to walking. This is accomplished by using sensors on the hip, which they relay information back to the CPU which determines how much assistance needs to be given. Using this assist the users will be able to walk for longer periods of time due to greater striding lengths(“Honda to…”, 2008). Related to the concept of technological development for SCI this device could serve as a continued rehabilitation mechanism that could be used in walking over ground as suggested by Winchester et al. (2005).

Honda also created a bodyweight support system which helps direct and control assistive force towards the user’s center of gravity while working in conjunction with the individual’s gait pattern. It uses a structure of a seat with a frame that attaches to the feet and supports the upward thrust needed when having to climb stairs or change bodyweight. Information on the walking / assistance patterns is determined by sensors within the shoes of the device. These sensors relay information to a pair of motions designed for the control of the right and the left legs(“Honda Unveils…”, 2008). This technology is specifically beneficial with regard to perception because it demonstrates collaboration between technology and the body in everyday use situations. It can help support the body on a sliding scale which can assist the individual as needed which is the basic platform that Edgerton and Roy (2009) suggest in their research.


Honda to Showcase Experimental Walking Assist Device at Barrier Free 2008. (2008) Retrieved January 15, 2009, from Honda Worldwide: Corporate. Web Site:

Honda Unveils Experimental Walking Assist Device With Bodyweight Support System (2008) Retrieved January 15, 2009, from Honda Worldwide: Corporate. Web Site:

Edgerton, V. R., Roy, R. R., & Larry, R. S. (2009). Sensorimotor Plasticity and Control of Movement Following Spinal Cord Injury +D18 Encyclopedia of Neuroscience, 629-635.

Taub, E., & Uswatte, G. (Writer) (1999). Constraint-Induced Movement Therapy: A New Family of Techniques with Broad Application [Article], Journal of Rehabilitation Research & Development: VA Prosthetics Research & Development Center.

Winchester P, McColl R, Querry R, Foreman N, Mosby J, Tansey K &Williamson J (2005). Changes in supraspinal activation patterns following robotic locomotor therapy in motor-incomplete spinal cord injury. Neurorehabil Neural Repair 19, 313–324.