Project Abstract

The retail industry faces significant challenges in managing stock efficiently and providing a seamless customer experience. Manual stocking and basket loading processes are labor-intensive, time-consuming, and prone to errors. In the rapidly evolving retail landscape, businesses are continuously seeking innovative solutions to enhance operational efficiency, optimize stock management, and elevate customer satisfaction. One promising approach is the integration of robotic assistants capable of aiding shopkeepers and customers in tasks such as stock arrangement and basket loading. This project aims to develop a cutting-edge vision system that will serve as a critical component of a robotic retail assistant, enabling accurate detection and recognition of merchandise items, efficient barcode scanning, and seamless integration with the overall robotic platform.Leveraging state-of-the-art computer vision techniques, machine learning algorithms, and barcode scanning technologies, the vision system will be designed to accurately identify and categorize various merchandise items, decode barcodes for pricing and inventory management, and provide real-time feedback to the robotic assistant. This will not only streamline stock arrangement processes but also enhance the customer experience by enabling efficient basket loading and checkout procedures.The project will encompass several key phases, including environment setup and data collection, object detection and recognition model development, barcode detection and decoding implementation, and extensive testing and validation in simulated and real-world retail environments. The modular design and implementation of the vision system will ensure scalability and adaptability, allowing for future enhancements and integration with additional functionalities or robotic platforms.By addressing the challenges of stock arrangement and basket loading, this project has the potential to revolutionize the retail experience, contributing to increased operational efficiency, improved inventory management, and elevated customer satisfaction. The successful deployment of the vision system and its integration with the robotic platform will pave the way for further advancements in robotic retail assistants, ultimately driving innovation and transforming traditional retail processes.

Keywords: Robotics, Data collection, Object detection

 Conference Details

Session: Presentation Stream 29 at Presentation Slot 8

Location: GH011 at Wednesday 8th 13:30 – 17:00

Markers: Alex Warren (GTA), Deshan Sumanathilaka (GTA)

Course: MSc Computer Science, Masters PG

Future Plans: I’m looking for an industry placement