Type of Submission
Poster
Proposal
The goal of this project is to design and build an inexpensive, voice-controlled prosthetic hand. Because precise electromyographic technology is often costly, our design utilizes speech recognition to move all five fingers of our bionic hand independently. This allows the user to command multiple different grips without the cumbersome and expensive equipment needed to detect individual finger movements.
Speech recognition is carried out by a hardware module, which can detect up to five unique voice commands at a time. To eliminate the chance of an inaccurately identified command, we installed an easy-access power button. In addition, an OLED (a digital screen) displays pertinent information about voice command sets and steps the user through re-training the speech module. Various buttons are installed to allow user input for this process. The bionic hand is controlled by an STM32 Blue Pill, which receives translated voice commands from the speech hardware module. To prevent servo damage when an object is grasped, pressure sensors made from a conductive material are attached to each finger on the prosthetic. The bionic hand and all circuit components are attached to a 3D printed socket, which can be molded to fit each user’s arm.
This entire system is designed with both cost and precision in mind. All components are inexpensive and easily replaceable, providing the user with an affordable prosthetic hand. However, while the parts are inexpensive, the system accurately detects voice commands in a consistent manner. In addition, the user interface is designed to be intuitive, allowing any user to easily train and use their own bionic hand.
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Voice-Activated Bionic Hand
The goal of this project is to design and build an inexpensive, voice-controlled prosthetic hand. Because precise electromyographic technology is often costly, our design utilizes speech recognition to move all five fingers of our bionic hand independently. This allows the user to command multiple different grips without the cumbersome and expensive equipment needed to detect individual finger movements.
Speech recognition is carried out by a hardware module, which can detect up to five unique voice commands at a time. To eliminate the chance of an inaccurately identified command, we installed an easy-access power button. In addition, an OLED (a digital screen) displays pertinent information about voice command sets and steps the user through re-training the speech module. Various buttons are installed to allow user input for this process. The bionic hand is controlled by an STM32 Blue Pill, which receives translated voice commands from the speech hardware module. To prevent servo damage when an object is grasped, pressure sensors made from a conductive material are attached to each finger on the prosthetic. The bionic hand and all circuit components are attached to a 3D printed socket, which can be molded to fit each user’s arm.
This entire system is designed with both cost and precision in mind. All components are inexpensive and easily replaceable, providing the user with an affordable prosthetic hand. However, while the parts are inexpensive, the system accurately detects voice commands in a consistent manner. In addition, the user interface is designed to be intuitive, allowing any user to easily train and use their own bionic hand.
