A Low Cost Upper Limb Exoskeleton for Rehabilitation and Arm Movement Assistance

Overview

More than 20 million Americans suffer from peripheral neuropathy, a disease that degrades both motor and sensory nerve control. This makes it almost impossible to perform vital daily tasks such as talking on the phone, or getting dressed everyday. Implementing an inexpensive, portable system that augments user motions and creates progress-based rehabilitation plans would be highly valuable. The purpose of this project is to investigate the feasibility of designing a low-cost, 3-D printed orthosis that augments patient movements and tracks rehabilitation progress. The system developed in this paper consists of a 3-D printed exoskeleton and 3 actuators; a 48mm stroke linear actuator, 2 360 degree rotational servo motors, and a control system that uses input from EMG sensors. The exoskeleton was modeled in a CAD program and 3-D printed out of ABS. It was placed on a 3-D printed arm that was modeled and printed out of PLA. A control system, which takes input signal from the EMG sensors, converts the electrical signal into a specific direction and motion for the actuator. Actuation on the initial prototype provides 2 separate degrees of freedom. The elbow is actuated with an electric linear actuator, which is attached to the 3-D printed orthosis above and below the joint. The abduction/adduction movement of the arm, which is actuated by a servo motor that rests on the top of the shoulder, is used to create a volume of space the patient can move in. Furthermore, shoulder flexion and extension increases the volume of motion created by the orthosis.The servo motor pulls bowden cable, which is attached to the orthosis near the top of the arm. EMG sensors, which are placed on the bicep, deltoid, and frontal deltoid, measure the strength of electrical activity in a muscle. These EMG sensors act as a user-input in the control loop, which determines the direction and speed of actuator movement. Data from the EMGs can also be stored over time to allow patients to track muscle rehabilitation. This paper experimentally confirmed the feasibility of creating a bio-inspired arm-assistive device that provides 3 degrees of freedom. The successful actuation of the shoulder joint gives patients a much larger range of motion compared to current exoskeleton arms/assistive devices and results demonstrate the system's ability to rapidly adjust actuation based on live input from EMG sensors. Furthermore, the entire control system developed in this project is able to detect user intent to move and augment this movement with accurate and precise actuators. Future work for this project include further optimizing shoulder actuator placement.