Co-Optimization of Robots and Mechanisms

Speaker: Andrew Spielberg
Affiliation: MIT

Abstract:  The design of robots and the design of robot motions are innately coupled problems. Changes in physical robot designs often require modifying or completely redesigning their motion primitives. Similarly, if trajectories that satisfy user constraints do not exist for a given robot in solving a required task, its mechanical design must be revisited.  As we attempt to transition our designs from the virtual world to the physical world, we encounter issues in fabricability, resource limitations, and ease of assembly.  In this talk, I present recent work (ICRA ’17) on the co-optimization of whole-body robot trajectories and design parameters from high-level task specifications.  Our method requires minimal user domain knowledge, requiring only a coarse guess of the target robot configuration sequence and a parameterized robot topology as input.  Further, our method considers the physical constraints afforded by the actuators, and selects the minimal actuators possible for task feasibility.  Following this work, I present methods for improving the fabricability of such mechanisms using cable-driven actuation (SCA ’17).  I then present work (IJRR ’17) which makes this otherwise mathematically-intensive design process more intuitive for human users by presenting more intuitive, end-to-end CAD-like robot design software.  Finally, I discuss what I believe are some of the most important open problems facing the automated robot design community moving forward.

Biography:  Andrew Spielberg is a third year PhD student in the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT. He is co-advised by Wojciech Matusik and Daniela Rus, working on the Printable Programmable Machines project. Prior to joining MIT, he earned his M.Eng in Computer Science and his B.S. in Computer Science and Engineering Physics at Cornell University, and then spent two years working as a research engineer in the Asymmetric Operations Department at the Johns Hopkins University Applied Physics Lab (JHU/APL). He has also participated in internships at Disney Research Pittsburgh and Disney Research Zurich.  Andrew’s work has been presented in the graphics, HCI, and robotics communities, and has been nominated for a best paper award at ICRA and earned a best paper award at CHI.

For more information, contact Prof. Ankur Mehta ()

Date(s) - Aug 03, 2017
11:00 am - 12:00 pm

E-IV Tesla Room #53-125
420 Westwood Plaza - 5th Flr., Los Angeles CA 90095