Professor Righetti takes his robotics research on the road

Professor standing next to a tall upright robot in a robotics lab

Associate Professor Ludovic Righetti in his Machines in Motion Laboratory

Associate Professor Ludovic Righetti’s Machines in Motion Laboratory is based right here at NYU Tandon, in Brooklyn, but he is frequently called upon to present his research at high-profile conferences and seminars around the world.

The latest presentation took place this month at Columbia University, where he spoke about his team’s groundbreaking work on nonlinear model predictive control (MPC), a rigorous and reliable technology used to get robots to walk and manipulate objects autonomously, and outlined their efforts to include multi-modal sensing and accelerate the generation of complex behaviors through machine learning and online optimization. 

Righetti, who holds an international chair at the Artificial and Natural Intelligence Toulouse Institute in France and serves as a Vice-President of the IEEE Robotics and Automation Society, is focused on developing algorithms to make robots that walk and manipulate objects autonomous, versatile and safe to interact with. His novel approach to machine learning and optimal control aims to create intelligent robots that “understand” when and how to interact with their environment, unknown objects, and people.

At Columbia, he explained that the algorithms designed at the Machines in Motion Lab are intended for real applications with the potential to make real social impact and discussed the responsibility robotics researchers have to ensure that those impacts are positive.

Just a few weeks before his presentation there, Righetti, an internationally recognized leader in legged locomotion research, had appeared at ICRA@40. This special conference celebrated the 40th anniversary of the IEEE International Conference on Robotics and Automation (ICRA), the world’s largest conference in robotics. Joining a slate that included prominent roboticists from across the field, he presented a talk that asked, “Is locomotion a solved algorithmic problem?” 

Righetti’s research has long been a regular feature of ICRA: in 2023, for example, he described a method whereby a robot is able to adapt to changes in the environment directly from sensor measurement, and the previous year, he had proposed a novel paradigm to incorporate explicit force sensing into a predictive controller.

Chances are good that when ICRA takes place in 2025, Righetti will once more be on the move.