OUR RESEARCH FOCUS
Getting robots to move in the real world is hard! The act of making or breaking contact or overcoming friction can cause the robot's state, contact forces, or control inputs to appear to change instantaneously in time. The resulting trajectories are nonsmooth and generating desired motions from these interactions are not trivial. However, these interactions are ultimately what make robotic systems valuable in the home, across industries, and in search and rescue operations.
Our goal is to address fundamental problems and design challenges in generating desired motions for robots with nonsmooth dynamics when getting themselves or an object from one location to the next.
OUR APPROACH
What insight can we gain from studying a robot's equations of motion? Are there particular motions that can serve as templates or primitives for generating a library of desired trajectories?
Most trajectory generation problems are search problems. How can we quickly and robustly converge to a desired trajectory when the robot is underactuated and the dynamics are nonlinear, nonsmooth, and nonconvex?
How does the theory and numerics hold up when applied to real systems? How can we design simple robots that capture the challenges of working with underactuated sytems while providing insight into the theory and numerics being developed?
Welome and Congrats to Parth Domadia on his RES-MATCH application. Parth will be working on our smart assistive device project.
Congrats to Amal Alwatik, Riki Osako, Vincent Trosky, and Ashfaq Siddiqui on their successful Armour R&D applications and poster presentations.
Amal Alwatik and Riki Osako presented their works at this year's Midwest Robotics Workshop.
Congrats to Aditi Kumar and Manav Tailor on their best poster awards in the College of Computing and Mechanical Engineering poster competitions.
Congrats to Amal Alwatik, Riki Osako, and Vincent Trosky on their successful Armour R&D applications.