Senior computer science and software engineering student programs independent AI robotic arm for everyday use
Published: Nov 19, 2024 1:30 PM
By Olivia Ballard
Imagine a robotic arm that can tell the difference between plastic fruit and real fruit.
Carson Bulgin, a computer science and artificial engineering senior, is at the forefront of artificial intelligence (AI) programming for assistive robots. He is developing software that enables a robotic arm to function without explicit supervision and human feedback.
Bulgin created a generalized framework to be integrated into the arm, a Dobot CR3, allowing it to recognize objects based on defining characteristics like shape, color and size. This will allow the arm to better understand surroundings and repeated actions.
By connecting application programming interfaces — the predefined functions that control movement and coordination — with a computer vision camera, the arm can recognize spatial awareness and objects.
“We want it to be able to fully recognize a wide variety of objects and figure out which one is best for a specific scenario,” Bulgin said. “Say you have a real pear and plastic pear next to each other. Without touching it, we want the arm to be able to figure out which is the fake and why.”
Bulgin’s project began with the end goal of helping people. Ultimately, he aims to develop advanced assistive robots for individuals with cognitive disabilities such as dementia, simplifying difficult tasks.
“The arm will be programmed to remember repeated tasks and items. For example, if someone forgets their keys a lot, the arm will remember the notion and automatically pick them up,” he said. “Another example is if someone is baking a recipe and they are getting ready to perform an action, the arm will pick up on that.”
Bulgin works in the Open-Ended Reasoning and Knowledge Acquisition (ORCA) Laboratory under Sathyanarayanan Aakur, assistant professor in computer science and software engineering. Aakur praises Bulgin’s project and developments within the lab.
"Carson’s project embodies our vision for computer vision and AI as tools that extend beyond recognition to active support and intuitive interaction,” Aakur said. “This work highlights the potential for AI-driven vision to empower individuals with cognitive challenges, fostering a new level of independence through adaptive, context-aware robotics."
Bulgin plans to continue his work with the arm and have a functional prototype by spring before he graduates. Next summer, he hopes to present his findings at the Institute of Electrical and Electronics Engineers and The Computer Vision Foundation Conference on Computer Vision and Pattern Recognition in Nashville, Tennessee.
Media Contact: , ccb0082@auburn.edu,
Carson Bulgin, computer science and artificial engineering senior, poses for a photo in a computer lab.
