Max Robotics — Global Service Robot Directory + US Certification

About

Soonyong Song is a leading researcher in the field of continual and lifelong learning for robotic vision. His work focuses on developing artificial agents that can learn continuously from their environment, much like humans, without forgetting previously acquired knowledge—a challenge known as catastrophic forgetting. Song’s major contributions include organizing and co-authoring the landmark “IROS 2019 Lifelong Robotic Vision: Object Recognition Challenge” (12 citations), which introduced the OpenLORIS-object dataset, a benchmark now widely used to evaluate lifelong object recognition in real-world scenarios. He also co-authored “The Present and Future of Continual Learning” (5 citations), a comprehensive survey that categorizes state-of-the-art approaches and outlines future directions for the field. Additionally, his report on the IROS 2019 Lifelong Robotic Vision Challenge (2 citations) details the methods and results from the top 8 finalists among over 150 teams, highlighting his role in advancing reproducible, competitive research. Song’s work is foundational for researchers aiming to build robust, adaptive robotic systems capable of operating in dynamic, open-world environments.