Cooperative Control Strategies for the Lokomat

Idea

The rehabilitation robot Lokomat has been developed at Balgrist University Hospital Zurich. It allows automated treadmill training for patients with motor impairments in the lower limbs. Originally, the Lokomat was only position-controlled. Thus, the patient’s legs were guided along a fixed reference gait pattern.

Patients can remain completely passive in this kind of training, which is not ideal for the rehabilitation process. Therefore, we aim at developing new control strategies, which promote active participation of patients during the training.

How Does a Cooperative Controller Work?

Ideally, the robot should provide as much freedom as possible to the patient and supports him or her just as much as needed. To optimize conditions for motor relearning, patients need to experience the results of their efforts and move as freely as possible – the robot needs to “get out of the way”. However, at the same time, the robot has to provide at least the amount of support that the patient needs to maintain stable walking.

Patient-cooperative robotic behavior is based on the following basic control strategies:

Impedance Control

The patient’s legs can deviate from the reference gait pattern, yet they are pushed back to the reference pattern by a virtual spring-damper element.

Path Control

This variant of impedance control limits the possible foot positions to a physio-logically meaningful path in space. The timing of the movement is left to the pa-tient. The effect can be imagined best as a virtual tunnel in space through which the patient is moving his or her leg.

Adaptive Control

We apply principles from adaptive control to determine how much support a patient needs and how he or she wants to move.

Where Are We Right Now?

We have finished a study addressing the question how cooperative robot-aided training influences the level of active participation of subjects with incomplete spinal cord injury. Together with our partners at Rehabilitation Institute of Chicago we will test the strategies also with stroke patients.

Current engineering research concentrates on improving the adaptive components of our cooperative strategies. Further studies will investigate, if increased patient activity due to cooperative training also translates into better rehabilitation outcome.

Contact Person

Please contact Alexander Duschau-Wicke for more information.

Financial Support

This project is part of the Machines Assisting Recovery in Stroke at the Rehabilitation Engineering Research Center (MARS-RERC) project supported by the U.S. National Institute on Disability and Rehabilitation Research (NIDRR), Grant Number #H133E070013. Further support is provided by the National Center of Competence in Research Neural Plasticity and Repair (NCCR Neuro), which is funded by the Swiss National Science Foundation (SNSF) and by Hocoma AG, Volketswil.