Portfolio

Project Description
The BoogieBot is a project designed over the course of 10 weeks, employing the ROBOTIS-OP3 humanoid robot to generate dynamic dance routines. This system analyzes music files to identify the track using shazamio and leverages Spotipy to determine the song's tempo. Utilizing this information, the BoogieBot choreographs dance sequences by aligning pre-designed movements to the rhythm of the music. The foundation of this system is built on a multi-docker framework, allowing for functionality across all components. Additionally, the project utilizes the ROS1 bridge, facilitating communication between the ROS1 and ROS2 frameworks.

Core Features:

Using shazamio and Spotipy:
The song artist and title are identified by inputting a .wav audio file into shazamio's `recognize_song` function. The song and artist are then utilized by the Spotipy `search` function, to retrieve the Spotify track id. Finally, the Spotify track id can be put into Spotipy's `audio_feature` function, which returns the song's tempo.

Docker Systems:
This project utilizes specialized Docker containers, designed to combine different ROS environments. The first container hosts ROS1 Noetic on Ubuntu 20.04, specifically chosen to satisfy a Python 3.8 requirement. The second container integrates ROS1 Noetic with ROS2 Galactic, operating on Ubuntu 20.04, allowing communication using the ROS1 bridge between different ROS versions. Both Dockers are able to communicate with eachother and the host computer. Custom Dockerfiles configure this system, ensuring that each component interacts smoothly within the ROBOTIS ecosystem. The use of Docker in this project is shown below.

ROS1 Bridge:
This project incorporates the ROS1 bridge, an essential component for communication between ROS1 and ROS2 systems. Deployed within a Docker container, the bridge is constructed using a multi-stage Dockerfile. The first stage creates the ROS1 environment (Noetic), the second stage creates the ROS2 environment (Galactic), and the third stage creates the ROS1 bridge environment (ROS2 Galactic), by combining the necessary components of the previous stages. The different stages allow for the ROS1, ROS2, and ROS1 bridge environments to be built properly. In this project, the bridge is used to send a custom message type from ROS2 Galactic to ROS1 Noetic.

State Machine:
Unless specified otherwise, all nodes are developed using ROS Kinetic and programmed in C++.

WAITING: By default, ROBOTIS remains idle until activated through a service call, initiating the transition to the next state.
PROCESSING: Handled by the 'audio' node, while in this state, ROBOTIS analyzes the audio file, identifying the track through shazamio and determining the song's tempo using Spotipy.
This particular node operates under ROS1 Noetic and is programmed in Python.
PICKING: Handled by the 'action_pick' node, here ROBOTIS selects dance movements that align with the song's tempo, determined by dividing the beats per minute (bpm) by 60.
This process takes into account the robot's physical limitations, and adjusts movements to align with the song's tempo.
This particular node operates under ROS2 Galactic.
EDITING: Handled by the 'action_edit' node, here ROBOTIS compiles the selected movements into a YAML action file, which contains all the details required for the execution of the dance routine.
PLAYING: Handled by the 'action_play' node, in this final state, ROBOTIS performs the choreographed dance.
(Note: The sound of the robot's motion may exceed the audio output; hence the overlay of music in the videos.)