Matthew Gombolay
Welcome!
I am a PhD candidate in the Interactive Robotics Group at MIT and have recently completed my Master's Thesis. After receiving a B.S. in Mechanical Engineering, I began work in the IRG developing methods for coordinating multi-agent teams (humans and robots) in time and space. We have applied this work to advanced robotic manufacturing of aerospace structures. At the same time, I think it is imperative that we understand the human factors issues in human-robot teamwork from a design perspective. With advanced algorithms and a proper understanding of human factors, I believe we can make the workplace safer, more efficient, and a more desirable place to work for humans.
Latest News
  • I just returned from UC Berkeley presenting my latest work on Human-Robot team coordination at Robotics: Science and Systems (RSS) 2014 at UC Berkeley. [36% Acceptance Rate]
  • I am now a Human-Robot Interaction Pioneer! My workshop paper was published in the proceedings of the 2014 HRI Pioneers Workshop from this year in Bielefeld, Germany. [36% Acceptance Rate]
  • My paper on Uniprocessor Scheduling was selected by the AIAA Intelligent Systems Technical Committee as the Best Intelligent Systems Paper from the AIAA 2012 Infotech@Aerospace Conference!
Dynamic Scheduling of Human-Robot Teams
New uses of robotics in traditionally manual manufacturing processes require the careful choreography of human and robotic agents to support safe and efficient coordinated work. Tasks must be allocated among agents and scheduled to meet temporal deadlines and spatial restrictions on agent proximity. These systems must also be capable of replanning on-the-fly to adapt to disturbances in the schedule and to respond to people working in close physical proximity.

 


We are developing fast, near optimal task assignment and scheduling algorithms that scale to multi-agent, factory-size problems and support on-the-fly replanning with temporal and spatial-proximity constraints. We demonstrate that this capability enables human and robotic agents to effectively work together in close proximity to perform manufacturing-relevant tasks, including multi-robot composite material placement, drilling, and robotic assistance in manual tasks. This research is performed in collaboration with Boeing Research and Technology.
  • Dynamic Environment requiring the ability to quickly adapt.
  • Computing the optimal schedule for large teams is computationally intractable.
  • Workers must remain at safe distances to prevent harm to humans or damage to robots.
Read more:
Human-Robot Interaction
I believe that it is essential we take a human-centered approach when integrating automation or robots into the workplace. With a background in Human Factors Engineering, I have begun work in understanding the impact of integrating robots into the manufacturing environment and how best to do so from a design perspective.

 


No matter how advanced or sophisticated the robotics or autonomy, if we do not design systems that are easy to use, understand, and appreciate from a human point of view, these algorithms will ultimately be unsuccessful.
  • Introducing robotic teammates into human domains such that the humans do not feel that their jobs are threated or that they are devalued.
  • Integrating advanced robot technology into the workplace safely and efficiently.
  • Building successful teams of humans and robots in terms of communication, shared mental models, and well-choreographed activity.
Read more:
Real-Time Processor Scheduling
Traditional scheduling in Artificial Intelligence (AI) AI and Operations Research (OR) often rely on methods for scheduling that are too slow for real-time, dynamic scheduling of large task sets. While real-time processor scheduling techniques are fast, they are often uninformative for more general, real-world problems. One goal of my research is to bridge the gap of these communities to provide methods for scheduling that are fast enough for real-time systems yet provide near-optimal solutions for real-world problems.
  • Handling task sets that have more general (i.e., real-world) constraints.
  • Generating near-optimal solutions for NP-Hard problems using fast (i.e. heuristic) methods
  • Bridge the gap between AI/OR and Real-Time Processor Scheduling by developing methods applicable to both fields of research.
Read more:
Selected Publications
Previous Work
  • M. Gombolay, S. Beder, R. Boggio, J. Samsundar, P. Stadter, and P. Binning. "Scheduling of Oversubscribed Space-Based Sensors for Dynamic Objects of Interest." Proc. of 9th Annual U.S. Missile Defense Conference and Exhibit, Ronald Reagan Building and International Trade Center, Washington, DC. March 2011.
  • T. Safko, D. Kelly, S. Guzewich, S. Bell, A. S. Rivkin, K. W. Kirby, R. E. Gold, A. F. Cheng, T. M. Aldridge, C. M. Colon, A. D. Colson, D. V. Lantukh, P. Pashai, D. Quinn, E. H. Yun, and the ASTERIA team. “ASTERIA: A Robotic Precursor Mission to Near-Earth Asteroid 2002 TD60.” Proc. of Lunar and Planetary Science Conference, Houston, TX. March 2011.