Live System Status

Smart LaundryManagement

Transform your laundry operations with real-time monitoring, intelligent notifications, and comprehensive analytics. Never miss a cycle again.

Explore Features

99.9%

Uptime

24/7

Monitoring

Real-time

Updates

Project Goals & Scope

Scalable, low-cost, and non-invasive monitoring.

LaundryIQ was built to modernize shared-laundry monitoring using vibration and temperature sensors to detect washer and dryer activity without modifying machines.

Core focus: live detection, uptime, and notifications.
Deployable across dorms, apartments, and shared facilities.
Non-invasive sensor integration with low maintenance needs.

LaundryIQ test setup in laundry room — Our physical setup during testing: sensor node on machines, local dashboard open, and utilities visible for context.

System Architecture

Hardware → Firmware → Cloud → Dashboard

Each ESP32-S3 device collects vibration and temperature data, processed through modular firmware before syncing with Convex. The Next.js dashboard visualizes machine states in real time.

Hardware Layer

ESP32-S3 with IMU and temperature sensors enables cycle state detection. Selected for built-in Wi‑Fi/BLE and headroom for lightweight AI analysis when needed.

Firmware Logic

Modular firmware separates sensing, networking, and OTA for easier debugging, plus board-specific headers and a central config make pin mappings/features easy to swap across revisions.

Web Dashboard

Next.js + Convex ensures fast, responsive real-time machine tracking.

Vibration Detection & Smoothing

Turning raw IMU readings into clear, robust cycle signals.

Why smoothing?

Raw magnitude often contains micro-spikes from incidental bumps and mechanical noise. A short moving average (and optional median pre-filter) stabilizes the signal so thresholds and hysteresis work reliably across machines.

How we use it

We compute accel magnitude, apply a windowed smoother (tuned empirically), then feed a debounced state machine: idle → running → spin → idle. This reduces false positives and improves cycle-boundary detection.

Side-by-side graph: left non-smoothed, right smoothed, showing clearer peaks after smoothing — Signal comparison: left = non-smoothed, right = smoothed. Smoothing amplifies the meaningful envelope and suppresses jitter, enabling stable thresholds.

AI Vision Integration

Computer vision enhances detection accuracy and user experience.

Our AI vision system provides an additional layer of intelligence, reading machine displays and detecting user presence to enhance the accuracy of cycle detection and provide richer status information.

Display Recognition

AI-powered computer vision reads the machine's display panel to show cycle status, time remaining, and machine state directly in the dashboard, providing users with detailed information beyond vibration detection.

Presence Detection

The vision system detects when users are present at machines, identifying when someone is removing clothes or loading laundry. This helps distinguish between completed cycles and active use, reducing false notifications.

Hardware Reliability

Power, layout, and materials decisions for real-world installs.

Power Choice

MVP uses external power to reduce heat/cost and improve uptime in warm, humid rooms.

PCB Layout Hurdles

Iterated routing to maintain antenna clearance and isolate noisy signals for stable RF and sensors.

Environmental Protection

Tested enclosure plastics for detergent, humidity, and heat resistance to preserve sensor accuracy.

Dashboard & Routing

Simple URLs, multi-site management, and scan-to-open access.

QR codes link directly to a machine's UUID page, so users scan to view live status instantly. Admins manage organizations and locations with readable slugs and guardrails that prevent collisions.

Human-readable org slugs for clean, shareable URLs.
QR codes generated to map each device to its dashboard.
Multi-location grouping to support enterprise deployments.

Security & Reliability

End-to-end encryption, OTA validation, and tamper resistance.

Security and reliability are built into every layer — from firmware authentication to encrypted API access and hardware tamper resistance. Endpoints follow a versioned API pattern and are validated with Postman collections to catch firmware↔web mismatches early.

Encrypted OTA and data transmission for every device.
Strict API key protection and device authorization.
Hardware-level tamper detection and safety layers.

Encrypted Communication

All telemetry and OTA requests are secured with device authentication.

Tamper Resistance

Hardware and software layers ensure data integrity and physical protection.

System Reliability

Resilient modular structure ensures continued uptime and stability.

A Student-Led Engineering Project

Self-directed development with structured project management.

LaundryIQ is entirely student-led, demonstrating our ability to manage a complex full-stack IoT project from conception to deployment. We created our own timeline, established clear milestones, and held ourselves accountable to deliver a production-ready system.

Self-Managed Timeline

We developed a comprehensive project timeline with clear milestones, breaking down the full-stack development into manageable phases from hardware prototyping to cloud deployment.

Task Ownership

Each team member took ownership of specific components, self-assigning tasks based on expertise and interest, ensuring accountability and efficient parallel development.

Progress Tracking

Regular check-ins and milestone reviews kept us on track, allowing us to adapt quickly when challenges arose while maintaining our delivery timeline.

Created detailed sprint plans with weekly goals and deliverables
Established clear ownership boundaries for hardware, firmware, cloud, and frontend components
Maintained accountability through peer reviews and shared documentation
Adapted our timeline dynamically when technical challenges required pivots
Delivered a fully functional MVP on schedule through disciplined self-management

Who We Are

Student-led full-stack IoT team.

Ethan Hen

Leeor Nahum

Elisabeth Yap

Contact the LaundryIQ Team

Pilots, partnerships, and feedback welcome. Reach out for demo access or collaboration opportunities.

Email Us See Demo