Smart Solutions Studio
The Smart Solutions Studio brings together like minded students who are interested in artificial intelligence (AI), edge computing, Internet-of-things (IoT) systems and general automation. Along with the studio supervisors, teams will be working together with industry/technology partners to understand the project needs and prototype solutions through a build, test and validation approach that is in line with the project partner's vision.
Project supervisors:
- Mr Eugene Ee (wheee@nus.edu.sg)
- Prof Chan Tong Leong (chan.tl@nus.edu.sg)
Studio timeslot in Semester 2 AY2025/2026:
- Thursday 1 pm to 4 pm
Examples of past projects:
This project aims to improve navigation for elderly and tech-averse visitors by developing an app-less smart wayfinding system that provides personalised, real-time guidance.
In this project, we designed VOTÉ, an AI-powered edge computer vision inspection system, to enhance quality control in Top-Level Assembly by providing real-time, step-by-step assembly guidance and preventing defects through customizable, no-code model training.
w[AI]ter: voice AI ordering and personalisation
w[AI]ter achieves higher customer engagement, personalised service, and accurate real-time responses to customer queries at full-service restaurants.
Our product, Eclat, enhances the experience of wine appreciators through a hassle-free, autonomous, and reliable household product that thoroughly washes and polishes premium wine glasses.
This project aims to enhance the accuracy of battery state-of-charge (SoC) and state-of-health (SoH) estimation for electric vehicles by developing a precise algorithm using Extended Kalman Filters (EKFs), targeting less than 2% margin of error.
Leak Hunter: non-invasive flow rate monitoring device for fluid pipelines
This project introduces a non-invasive device to monitor single-phase fluid flow rates in pipelines, alerting users when critical thresholds are reached for improved decision-making and operational efficiency.
VisionAI
This project aims to simplify the workflow of training and deploying AI models, specifically computer vision models to edge devices/ embedded systems.
AIoT predictive maintenance solutions for water pumps
Our project aims to design a predictive maintenance solution for water pumps utilising IoT, AI, and edge computing technologies.
Autonomous solution for blood transportation in hospital
In this project, we designed an autonomous solution to offer hospitals a streamlined and secure blood transportation solution, enhancing efficiency, saving lives in emergencies, and minimizing the reliance on manual labour.
EcoAir: IoT-enabled energy consumption monitoring in warehouses
For those managing large, cooled spaces who want higher efficiency, EcoAir provides an easy-to-install, retrofitted solution which can reduce the energy consumption of your cooling system while being significantly cheaper than hardware upgrades.
Go deep in green with Bang & Olufsen
This project aims to revolutionise the consumer electronics industry, specifically Bang & Olufsen, by designing both a sustainable packaging and an eco-friendly battery source, ensuring a greener and more environmentally responsible future for tech products.
Voice detection by bone conduction
In this project, we designed an earbud with the optimal bone conduction microphone position which aims to help users "discover their voice, amidst the noise".
Agentic AI to augment hospital workforce
This project aims to develop an AI agent alongside a vision system (camera) that aims to assist nurses in wards that have a high patient-to-nurse ratio. The AI agent is targeted to perform constant monitoring of patients and alert nurses through audio or visual cues should any of their patients be at risk.
Project partners:
- Problem owner: Alexandra Hospital
- Potential technology partner: Venture Corporation
AI operator assembly monitoring
Manual assembly errors often lead to costly material waste, rework labor, and production bottlenecks. This project aims to develop an AI toolkit that is able to monitor and validate operator actions in real time to prevent errors before they happen.
Project partners:
- Problem owner: TE Connectivity
- Potential technology partner: Advantech