Mechatronics Resarch Lab News
- All
- Presentations
- News Reports
- Conference

Congratulations Xiaotong!
Here’s to Xiaotong’s successful thesis defense! His presentation of “Relevance for Human-Robot Collaboration: Definitions, Systems, Algorithms, and Applications” has earned him his PhD in Mechanical Engineering. Congratulations on all the hard work and time he put into his work, and we all wish him good luck with future ventures!

MRL Says Goodbye to Dr. Fangzhou Xia
Our current research scientist Dr. Fangzhou Xia said goodbye to the Mechatronics Research Lab this weekend at a farewell dinner, celebrating his hard work at MIT for the past nine years.
Fangzhou will begin his journey away from MIT as a professor in mechanical engineering at the University of Texas at Austin. We wish him the best of luck in his future professional endeavors!

Qatari University Visits MIT
This summer, two visitors from the University of Doha visited MIT, including President of the university, Dr. Salem Al-Naemi, and Dr. Rachid Benlamri, the Vice President of Academics. Kamal and his guests spoke to innovation hubs on campus, hoping to facilitate the beginning of a collaboration between the two institutions. Among these offices were MIT's Venture Mentoring Service (VMS), the Kendall Square Initiative, and Office of Strategic Alliances and Technology Transfer (OSATT).

Congrats Class of 2024!
We're thrilled to announce the graduation of several members from Professor Kamal Youcef-Toumi's Mechatronics Research Lab! Among them are undergraduate, graduate, and doctoral students, all of whom have demonstrated exceptional potential in the field of mechanical engineering.

JSW Visits MIT
One of the Mechatronics Research Lab's current sponsors and collaborators, Japan Steel Works (JSW), came to visit MIT this spring where they toured several labs on campus. Current MIT students Andrew Palleiko and Jiajie Qiu and visiting scholar Jun Suzuki showed off the testbed in the lab!

Recent Publication on Invertible Symbolic Regression
This innovative method identifies analytical relationships between the inputs and outputs of a given dataset using invertible maps. By combining the concepts of Invertible Neural Networks (INNs) and Equation Learner (EQL), ISR creates an end-to-end differentiable invertible symbolic architecture, enabling efficient gradient-based learning.
One of the key features of ISR is its versatility. It serves as a normalizing flow for density estimation and is effective in solving inverse problems. This work represents a significant step towards developing interpretable, symbolic invertible architectures.

MRL Wins Best Annual Paper Award
Xiaotong Zhang's joint publication with Prof. Kamal was honored with the 2023 Best Paper Award from the esteemed IEEE/ASME Transactions on Mechatronics.
Their paper pioneered a magnetohydrodynamics energy harvester, using permanent magnets and water as a conductor to power wireless sensor networks. Their innovation has the potential to solve the power delivery bottleneck commonly encountered in wireless sensors and robots.