What it does
A smart carpet cleaner that uses AI vision (conceptual) and smart sensors to detect stains, map cleaning paths, and automate deep cleaning with minimal human effort — solving the problem of inefficient and labour-intensive traditional carpet cleaning.
Your inspiration
The idea came from observing how existing robotic cleaners fail to address the unique challenges of carpet cleaning — particularly stubborn stains and varying carpet types. Inspired by the growing demand for smarter home appliances and our own experiences with ineffective carpet cleaners, we saw the opportunity to design an automated solution that adapts to different carpet surfaces, improves cleaning efficiency, and reduces user workload. A market survey we conducted confirmed this gap, especially in households and commercial spaces with large and carpeted areas.
How it works
Our smart carpet cleaner is an automated cleaning device designed specifically for carpet surfaces. It uses smart sensors and rotating brushes to identify dirty areas and clean them with minimal human effort. While full AI functionality is planned for the future, the current prototype uses an ultrasonic sensor to detect obstacles and a rain sensor to ensure water does not enter the electronics. These allow the cleaner to navigate around furniture and stop if there is a safety issue. The core cleaning mechanism includes two circular nylon brushes, which rotate to remove stains and embedded dirt. Unlike vacuum cleaners that use suction, our brush system is quieter and more effective on carpets, especially for deep cleaning. The device is powered by a rechargeable battery and controlled via an Arduino Mega 2560 microcontroller, which manages movement, sensor input, and brush control.
Design process
Problem Research: We conducted surveys and analysed feedback from homeowners and commercial users. Most found current cleaners too manual, noisy, or ineffective on carpets. Idea Generation: We brainstormed multiple design concepts and selected one using a concept matrix based on cleaning effectiveness, size, cost, and user-friendliness. CAD Modelling: We created detailed 3D models of both a prototype and an ideal version using SolidWorks. Prototyping: We fabricated the cleaner using a plywood base (lightweight and strong) and an acrylic cover (transparent for easy inspection). All components were selected for affordability and availability. Testing and Simulation: We tested movement, brush efficiency, and structural durability. Engineering analyses (vibration, stress, and motion) were done using MATLAB and ANSYS to ensure stability and safety.
How it is different
Unlike most robotic cleaners that are built mainly for hard floors, our design is carpet-specific. It offers: Stain-focused cleaning via smart detection, Dynamic brush control to adapt pressure and speed, A modular design for easy maintenance and repair, A quiet operation by avoiding noisy suction systems. Additionally, there is no existing patent in Southeast Asia for such a device, highlighting its market opportunity.
Future plans
We aim to fully integrate AI features using a Raspberry Pi and HD camera for real-time image analysis. This will allow: Automatic detection of stain types, Mapping the cleaning path intelligently, Adapting cleaning intensity on the spot. We also plan to develop: Mobile app and voice control, Two product tiers (basic and premium) to meet different budgets, A more compact version for residential use and a larger version for hotels and offices.
Share this page on