Prof. Sertac Karaman
Professor of Aeronautics and Astronautics
Director, Laboratory for Information and Decision Systems (LIDS)
Primary DLC
Department of Aeronautics and Astronautics
MIT Room:
32-D784
Areas of Interest and Expertise
Autonomous Robotic Systems
Statistical Mechanics
Statistical Mechanics
Research Summary
Professor Karaman's primary research interests lie in the broad area of analysis, design, and optimization of provably-correct computationally-efficient planning and control algorithms for safety-critical cyber-physical systems, with a large emphasis on robotics applications. Work usually involves a rigorous theoretical analysis that is tightly coupled with implementation and experimental validation on a variety of robotic platforms.
Over the past four years, Karaman has worked on numerous different projects with topics ranging from motion planning to robust control theory and applications ranging from ground robotics to nano-imaging. Currently, he is very engaged with the rigorous analysis of fundamental properties of sampling-based motion planning algorithms. Karaman is also interested in extending the application domain of these algorithms to novel directions such as motion planning with complex task specifications, differential games, stochastic optimal control, and optimal estimation. Most recently, he has been working on the analysis of high-performance navigation through a randomly-generated cluttered environment, motivated by birds’ flight through a dense forest.
Anytime sampling-based algorithms for control problems
Anytime sampling-based algorithms with suitable convergence guarantees, such as the RRT*, hold the potential to become a practical solution to many problems in robotics and control theory in a unifying way. http://sertac.scripts.mit.edu/web/?p=502
High-speed navigation in cluttered environments
On one hand, roboticists have long been working on agile robotic vehicles that can quickly navigate through cluttered environments. On the other hand, biologists have been trying to understand how some birds manage to fly through dense forests at amazingly high speeds. For instance, a goshawk can reach speeds up to 40 miles per hour while flying through woodlands, according to a BBC documentary. Motivated by these exciting applications in robotics and biology, we analyze high-speed navigation through clutter from a theoretical standpoint by establishing novel connections with statistical physics.
http://sertac.scripts.mit.edu/web/?p=528
Over the past four years, Karaman has worked on numerous different projects with topics ranging from motion planning to robust control theory and applications ranging from ground robotics to nano-imaging. Currently, he is very engaged with the rigorous analysis of fundamental properties of sampling-based motion planning algorithms. Karaman is also interested in extending the application domain of these algorithms to novel directions such as motion planning with complex task specifications, differential games, stochastic optimal control, and optimal estimation. Most recently, he has been working on the analysis of high-performance navigation through a randomly-generated cluttered environment, motivated by birds’ flight through a dense forest.
Anytime sampling-based algorithms for control problems
Anytime sampling-based algorithms with suitable convergence guarantees, such as the RRT*, hold the potential to become a practical solution to many problems in robotics and control theory in a unifying way. http://sertac.scripts.mit.edu/web/?p=502
High-speed navigation in cluttered environments
On one hand, roboticists have long been working on agile robotic vehicles that can quickly navigate through cluttered environments. On the other hand, biologists have been trying to understand how some birds manage to fly through dense forests at amazingly high speeds. For instance, a goshawk can reach speeds up to 40 miles per hour while flying through woodlands, according to a BBC documentary. Motivated by these exciting applications in robotics and biology, we analyze high-speed navigation through clutter from a theoretical standpoint by establishing novel connections with statistical physics.
http://sertac.scripts.mit.edu/web/?p=528
-
Projects
January 22, 2019Department of Aeronautics and Astronautics
Fast Frontier-Exploration for Unmanned Autonomous Vehicles with Resource Constraints
Principal Investigator Sertac Karaman
January 22, 2019Department of Aeronautics and AstronauticsEnergy-Efficient Deep Neural Network for Depth Prediction
Principal Investigator Sertac Karaman
January 25, 2017Department of Aeronautics and AstronauticsPractical Algorithms and Fundamental Limits for Complex Cyber-Physical Systems
Principal Investigator Sertac Karaman
September 13, 2016Department of Aeronautics and AstronauticsA Parallel Autonomous Driving System
Principal Investigator Sertac Karaman
May 5, 2016Department of Aeronautics and AstronauticsHigh-Speed Flight In a Poisson Forest
Principal Investigator Sertac Karaman
May 5, 2016Department of Aeronautics and AstronauticsFunction-Train: A Continuous Analogue of the Tensor-Train Decomposition
Principal Investigator Sertac Karaman
-
Video
4.5.23-AI-Karaman
Future of Aerial Vehicles Enabled by Agile Autonomy 4.5.23-AI-Panel-Discussion
MIT and Industry Panel Discussion: The Present and Future of AI and Autonomy RD-11.15-16.2022-Karaman
Sertac Karaman
Director of the Laboratory for Information and Decision Systems (LIDS)
Associate Professor of Aeronautics and Astronautics at the Massachusetts Institute of Technology, Laboratory for Information and Decision Systems (LIDS)5.5.22-Efficient-AI-Sertac-Karaman
Sertac Karaman
Associate Professor, Aeronautics and Astronautics2020 Autonomy Day 1 - Sertac Karaman
As the technology for autonomous vehicles matures, the broad reach of the technology comes to focus, together with the new challenges and the shifting opportunities. The car that can drive itself under any condition better than any human driver - the holy-grail of autonomous vehicles - may not be as close as once thought. However, it is becoming clear that other opportunities with tremendous economic and social impact may be well within reach. In fact, fielding autonomous vehicles on the ground, in the air, on the water and even in space may transform a number of existing industries and create new ones. In this talk, we discuss three emerging technologies that will allow autonomous vehicles to interact with humans, rapidly react to their environment, and showcase complex autonomy even in miniature form factors, respectively. We also briefly discuss opportunities in business and in teaching of autonomous vehicles.