Design and testing of a physical therapy device controlled with voice commands

Publication Date


Document Type

Conference Proceeding

Publication Title

Proceedings of the ASME 2020 International Mechanical Engineering Congress and Exposition. Volume 5: Biomedical and Biotechnology






The world population is aging. Age-related disorders such as stroke and spinal cord injury are increasing rapidly, and such patients often suffer from mobility impairments. Wearable robotic exoskeletons are developed that serve as rehabilitation devices for these patients. An assistive knee brace is a simple wearable exoskeleton which is used to help people with mobility issues. This device provides partial assistance to the user and also helps in providing locomotion. Many exoskeletons are currently available in the market that have different functions and use. It is believed that, to date, no voice-controlled knee brace exists in an orthotic application, and that this project debuts a unique approach. This project presents the design of an assistive bionic knee joint with a motor-based actuator. The new exoskeletal mechanism uses the serial elastic actuator concept and mainly consists of a stepper motor, a ball screw, a set of spur gears, and a set of linear springs. The ball screw provides a linear movement to mimic the stretching and retracting action of a human knee. To create a proof-of-concept of the design, 3D printing is used. A voice recognition system has been developed in-house to control the exoskeleton using very simple voice commands. The motor is controlled using a motor driver and powered using an external power source. The 3D printed prototype with integrated voice-control module is tested for its essential functions. The test setup is loaded on the leg of a mannequin and tested under both no-load and full-load operation. The concept is proven to be successful in providing assistance to the human knee. However, the 3D printed material is observed to be bending, causing disruptions in the device's operation. The reaction times are expected to be significantly larger compared to the theoretically calculated values.

Funding Sponsor

San José State University


Assistive joint, Rehabilitation, Robotics, Voice control


Mechanical Engineering