Mapache

Mapache is a web application that visualizes environmental sensor data collected by RaccoonBot, an autonomous environmental monitoring robot created by the UC Irvine Robot Ecology Lab and deployed at the Crystal Cove Conservancy, CA

Link to Web App: https://raccoonbot---mapache.web.app/ (Currently, only data from May 6th to May 9th 2025 is available, while waiting for hardware updates on the RaccoonBot)

Overview

Mapache serves as the frontend interface in an Edge-to-Cloud (IoT) architecture. It is designed to provide researchers and field operators with live insights into environmental conditions during outdoor robot deployments, as well as export and data download capabilities

System Architecture

1. Sensor Collection (Edge Layer)
   The RaccoonBot carries sensors (temperature, humidity, pressure, Co2, TVOC and AQI) and uses an onboard microcontroller to collect readings in the field.

2. Wireless Transmission
   The microcontroller transmits data via a LoRa module to a nearby base station server on a Raspberry Pi.

3. Edge Gateway Processing
   The Raspberry Pi server acts as a relay, receiving LoRa signals and forwarding the structured sensor data to the cloud database.

4. Cloud Database Storage (Cloud Layer)
   The data is securely stored in a Firebase Firestore database, enabling real-time access and historical tracking.

5. Frontend Interface (Mapache)
   Mapache connects directly to Firestore to display historic and real-time sensor readings. It presents the data in a clean, responsive interface suitable for both desktop and mobile use.

Features

• Real-time data syncing from Firestore
• Organized display of sensor values with timestamps
• Easy to deploy and maintain

Tech Stack

• React + TypeScript
• Firebase Firestore (NoSQL, real-time database)
• Static hosting (Firebase Hosting)
• LoRa + Raspberry Pi used for backend transmission (in separate repo)

Repository Scope

This repository contains only the frontend client of the Mapache project. The backend logic (Raspberry Pi and microcontroller firmware) is managed separately.

Future Work

• Integration of machine learning models for data trend detection and acquiring ecological insights
• Enhanced charting and graphing functionality

Author

Agaton Pourshahidi
Robot Ecology Lab, UC Irvine
ajpoursh@uci.edu