Cross-Team Analysis Toolbox

Team Members

Ajay Parthibha, Amelia Zheng, Rhea Agrawal

Summary

The Cross-Team Analysis Toolbox is an interactive visualization platform designed to support the Mechanical, Electrical, and Autonomy subteams in analyzing vehicle performance data.

It provides two core functionalities:

• Dynamic Analysis: Visualizes recorded ROS bag data
• Static Analysis: Parses and visualizes CSV and ROS bag data, creating dynamic plots

The project is fully dockerized for easy deployment and can be hosted locally within the ELL environment.

Design Flow

User uploads ROSBag or CSV files via the web interface. Files are parsed and stored in PostgreSQL, and available for users to select from a dropdown menu. Workflows are divided into:

• Dynamic Analysis → ROSBag replay and 3D visualizations
• Static Analysis → CSV data plots and statistics

How to Use

1. Clone the repository
2. Run docker-compose up --build

Application Architecture

Frontend

• React.js: Displays visualizations, manages user interactions, and communicates with backend APIs.
• Shiny for Python: Provides the static visualization interface for CSV and parsed ROSBag data### Backend Python, Node.js, Express.js, PostgreSQL

Backend

Python (PyWorker):

• Parses CSV and ROSBag data.
• Handles database uploads and data formatting

Node.js + Express.js:

• Manages routing and server endpoints.
• Interfaces with PostgreSQL for data retrieval

PostgreSQL:

• Stores parsed CSV and ROSBag data

Usage