Sensor Finger

Description

The primary objective of this project was to design and develop a sensory glove capable of accurately translating hand movements into digital signals. Due to the complexity involved in creating detailed sensor mechanisms for each finger, the scope was initially focused on developing a single finger prototype, which could subsequently serve as a template for duplicating additional fingers.

The glove component was meticulously designed from scratch and produced using 3D printing technology with resin materials. This design specifically facilitated precise measurement of joint angles in the finger. Accompanying the glove was a custom-designed wrist-mounted device, integrating an STM32 microcontroller, tasked with capturing sensory data from the glove and transmitting this information to a Raspberry Pi single-board computer via the CAN communication protocol.

The Raspberry Pi handled data processing and relayed processed data to a PC utilizing the zeroMQ messaging protocol. Visualization of glove movements was implemented through the MuJoCo simulation environment, employing a 3D mesh model generated from the glove’s original CAD design.

Due to confidentiality agreements with our industry partner, the complete source code for this project remains private. However, the visualization component is publicly accessible through Gitlab. Additionally, comprehensive project documentation, although in Polish, can be accessed here and translated via services such as Google Translate.

This project was made for my university course called “Collaborative Project” at Wroclaw University of Science and Technology in collaboration with Clone Robotics