Data-Driven Control of an Ultracompact Industrial Robot

In recent years, automation has become more accessible to small- and medium-sized businesses, leading to an increase in popularity of ultra-compact and easy-to-integrate industrial robot arms like Mecademic’s Meca500. However, because of their size constraints, it is harder for these robots to accurately follow a programmed path. This research project aims to improve the path-tracking performance of Mecademic’s Meca500 robot by fusing state-of-the-art machine learning techniques with modern control design techniques. Improving the path-accuracy of the Meca500 will strengthen Mecademic’s competitive advantage in the fast-paced industrial automation market.

Faculty Supervisor:

James Richard Forbes

Student:

Steven Dahdah

Partner:

Mecademic

Discipline:

Engineering - mechanical

Sector:

Manufacturing

University:

McGill University