Individuals with stroke often experience substantial muscle weakness that greatly reduces their quality of life and leads to additional long-term health problems. Rehabilitation using robotic devices is a promising approach to intensify therapy and to expedite the recovery process after stroke. However, there still exist many barriers that limit the usability of robotic devices in the clinic or the patient's home. Key barriers include the high cost of these devices, their size, and their weight. Furthermore, safety can be a concern, especially when the devices are operated without the supervision of a therapist. Consequently, most clinics and patients still rely primarily on simple devices, such as weights, pulleys, and elastic bands, for performing functional exercises to regain muscle strength and movement control during rehabilitation. This project seeks to investigate a new class of robotic devices that has the potential to be inexpensive, small, and inherently safe such that robotic therapy can be translated into small clinics and even patients' homes. The key innovation is the use of computer controllable brakes, instead of motors, to generate programmed forces to resist or gently guide the user's motion during training. Compared with motors, brakes are lighter, smaller, cheaper, and most importantly, inherently safer because they can only remove energy. Thus, brakes will never hurt a patient, even in the case of a malfunction. At the same time, the devices will maintain many of the advantages of motorized robotic systems, such as the ability to guide the patient's movements along desired paths, to track progress and scale training intensity for adequate progression during rehabilitation, and to enable interactive "game"-like activities. These features could substantially increase the amount of therapy that patients can receive while making training fun and engaging. This project will establish the basic design and control principles for such robots, develop two prototype robots, and then validate the clinical potential of these novel devices in healthy subjects and stroke survivors. The project will advance the field of rehabilitation robotics by creating the scientific foundation for a new type of rehabilitation robot that will bridge the gap between fully assistive motorized robots and traditional exercise equipment. Project outcomes may not only impact the lives of millions of stroke survivors, but also others with a wide range of neurological or orthopaedic conditions, such as spinal cord injury, cerebral palsy, and joint surgeries. Additionally, the research will train students and scholars and foster outreach activities through educational opportunities for underrepresented individuals in Science, Technology, Engineering, and Math.
This project focuses on investigating the potential of a new class of semi-passive rehabilitation robots that will bridge the gap between current effective robotic devices that are too expensive and bulky for in-home rehabilitation and less effective, inexpensive, and totally-passive devices such as weights and elastic bands. The system will use controllable brakes instead of motors to generate forces to resist or gently guide the user's motion during training, which makes the device cost-effective, portable, and inherently safe for in-home use. The Research Plan is organized under three tasks. The first task is to systematically investigate how to design and control semi-passive rehabilitation robots. The primary challenges to overcome are that such robots cannot produce arbitrary forces, but can only create forces that remove energy from the device, and that energy dissipation is in the joints of the robot, not in the robot's end-effector. Thus task activities include: 1) systematically investigating different robot structures and other design choices that influence how the motion of the end-effector handle is translated to the joints and brakes and 2) developing controllers that allow the generation of force-fields, virtual walls, other haptic elements and closed-loop steering in trajectory tracking tasks. The second task is to conduct a thorough evaluation of the proposed device's capabilities. Task activities include: 1) developing hardware and software for prototype robots and 2) conducting tests that range from physical assessments to experiments with healthy human subjects, e.g., measuring how different types of resistance affect muscle activation of healthy subjects and implementing resistive and assistive capabilities of the robot in a version of a computer game--Pac-Man--to demonstrate how interactive elements can be used in a rehabilitation setting. The third task is to evaluate the clinical potential of the proposed device in stroke survivors. Task activities include: 1) testing how different types of resistance affect muscle activation in stroke survivors, 2) evaluating the extent to which closed-loop controllers improve performance of line-drawing tasks and 3) assessing the device's ability to break dysfunctional synergies and how training with the device affects movement kinematics, muscle coordination, and motor cortical plasticity during a simple reaching task.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
