Embedded Files
Overview
Carnegie Mellon Robotics Institute - Bot Intelligence Group
June 2023 - Dec 2024
Pittsburgh, PA
My Achievements:
Developed an innovative method for acoustic sensing to enhance 3D proprioception in soft robotics.
Designed and molded a soft robot, specifically tailored to incorporate audio components, functioning in tandem with a custom-designed audio PCB.
Created a 3D-printed apparatus to encase the robot, facilitating reliable testing and experimentation.
Wrote a series of Python scripts to control the soft robot's movements, transmit and record audio signals, produce intricate point-clouds, and streamline the data collection process.
Trained a neural network to accurately estimate the robot's position in real-time.
This position initially began in-person at Carnegie Mellon University, before transitioning to a virtual format for the remainder of 2023.
Research Paper Published to the IEEE International Conference on Robotics and Automation in PACIFICO Yokohama as the 2nd Author.
Presented my work at the annual Carnegie Mellon Robotics Institute Summer Scholars showcase.
SOFT ROBOT COMPONENTS
SOFT ROBOT COMPONENTS
The soft robot utilized a Raspberry Pi for data (audio) collection and an Arduino Nano for audio signal production. A modified 6-microphone array PCB was utilized to receive the audio signals.
ACOUSTIC PROPRIOCEPTION MODEL TRAINING
ACOUSTIC PROPRIOCEPTION MODEL TRAINING
A 3D camera was used to train the position estimation model with over 200 different positions.
MOTION TEST
MOTION TEST
PCB MODIFICATION AND WIRING
PCB MODIFICATION AND WIRING
MICROPHONE INTEGRATION
MICROPHONE INTEGRATION
AcousticSensing_SoftRobot_ZivenLopez.pdfLopez_Ziven_SoftRobotProprioception.pdfAcoustic Soft Robotic Proprioception for Omnidirectional End-effectors.pdfPage updated
Google Sites
Report abuse