Best Paper Award at CIE51, 2024

The Industrial AI Group celebrates Ilango Kandasamy and Yuqian Lu's Best Paper Award at CIE51 (December 9–11, 2024) at UNSW Sydney for their work on "Instructing Collaborative Robots using Large Language Models for Human-Robot Collaboration."

Kandasamy recently completed his Master of Robotics & Automation Engineering program.

The research addresses human-robot interaction challenges in smart manufacturing. The team explored using Large Language Models to interpret natural language assembly instructions and convert them into executable robot tasks.

Their work evaluated GPT-3.5's performance in semantic parsing using the HA-ViD dataset, demonstrating strong generalisation for previously unseen industrial assembly instructions through prompt-based learning.

The study additionally investigated Vision Language Models for object detection and proposed an end-to-end system architecture illustrating the integration of voice and LLMs, demonstrating GPT-3.5's performance in converting natural language instructions into low-level robot actions for industrial assembly tasks.