Dolly : a voice-activated and gesture-recognizing smart trashcan utilizing the YOLO alogorithm for autonomous navigation / Gerald M. Carique, John Mark C. Ruam, and Arnold Rie J. Sejera

By: Contributor(s): Material type: TextTextDescription: 135 leavesContent type:
  • text
Media type:
  • unmediated
Carrier type:
  • volume
Dissertation note: Undergraduate Thesis (BSCS) - Computer Science. Camarines Sur Polytechnic Colleges, 2024. Abstract: Effective waste management is a growing global concern, with an estimated 2.24 bil-lion tons of solid waste generated annually. For individuals with mobility impairments, disposing of waste independently remains a significant challenge. This study addresses this gap by developing "Dolly", a voice-activated, gesture-recognizing smart trashcan designed to enhance waste disposal efficiency and accessibility. Using the YOLO (You Only Look Once) algorithm, the system detects a thumbs-up gesture to trigger autonomous navigation, allowing the trashcan to move toward the user's location. This innovation is particularly useful for office workers and individuals with disabilities, promoting inclusivity and sus-tainability. The study's objectives include training and evaluating the YOLO algorithm for gesture recognition, integrating the model into a smart trashcan prototype, and testing the prototype's effectiveness in real-world scenarios. System integration used a Raspberry Pi 4 and Arduino Uno, and the system achieved a mean average precision (mAP@95) of 93%. Voice commands were incorporated through the PICO Voice Porcupine Wake Word engine. Results show that "Dolly" significantly improves accessibility for users with mobility im-pairments, promoting independence in waste disposal.
Tags from this library: No tags from this library for this title. Log in to add tags.
Holdings
Item type Current library Shelving location Call number Copy number Status Date due Barcode
Unpublished Materials Unpublished Materials Main Library Undergraduate Thesis Section UTH BSCS C277d 2024 (Browse shelf(Opens below)) 1-1 Not for loan 007015UM

Undergraduate Thesis (BSCS) - Computer Science. Camarines Sur Polytechnic Colleges, 2024.

Includes bibliographical references.

Effective waste management is a growing global concern, with an estimated 2.24 bil-lion tons of solid waste generated annually. For individuals with mobility impairments, disposing of waste independently remains a significant challenge. This study addresses this gap by developing "Dolly", a voice-activated, gesture-recognizing smart trashcan designed to enhance waste disposal efficiency and accessibility. Using the YOLO (You Only Look Once) algorithm, the system detects a thumbs-up gesture to trigger autonomous navigation, allowing the trashcan to move toward the user's location. This innovation is particularly useful for office workers and individuals with disabilities, promoting inclusivity and sus-tainability. The study's objectives include training and evaluating the YOLO algorithm for gesture recognition, integrating the model into a smart trashcan prototype, and testing the prototype's effectiveness in real-world scenarios. System integration used a Raspberry Pi 4 and Arduino Uno, and the system achieved a mean average precision (mAP@95) of 93%. Voice commands were incorporated through the PICO Voice Porcupine Wake Word engine. Results show that "Dolly" significantly improves accessibility for users with mobility im-pairments, promoting independence in waste disposal.

There are no comments on this title.

to post a comment.