Machine Learning Helps Robot Swarms Coordinate
Soon-Jo Chung, Bren Professor of Aerospace, Yisong Yue, Professor of Computing and Mathematical Sciences, postdoctoral scholar Wolfgang Hönig, and graduate students Benjamin Rivière and Guanya Shi, have designed a new data-driven method to control the movement of multiple robots through cluttered, unmapped spaces, so they do not run into one another. "Our work shows some promising results to overcome the safety, robustness, and scalability issues of conventional black-box artificial intelligence (AI) approaches for swarm motion planning with GLAS and close-proximity control for multiple drones using Neural-Swarm," says Chung. [Caltech story]
Myra Cheng Receives Goldwater Scholarship
Myra Cheng, an undergraduate student in computing and mathematical sciences, has been selected to receive a Goldwater Scholarship. The Barry Goldwater Scholarship and Excellence in Education Program awards scholarships to college sophomores or juniors who intend to pursue research careers in science, mathematics, and engineering. Myra works with Yisong Yue, Professor of Computing and Mathematical Sciences, and Joel Burdick, Richard L. and Dorothy M. Hayman Professor of Mechanical Engineering and Bioengineering; Jet Propulsion Laboratory Research Scientist, on optimization algorithms that can learn individual preferences based on real-time interaction with people. These algorithms can be used in wearable exoskeletons that help mobility-impaired individuals walk. "I'm interested in how machine learning interacts with humans and, more broadly, human society," she says. Cheng has also been working with Katie Bouman, Assistant Professor of Computing and Mathematical Sciences, Electrical Engineering and Astronomy; Rosenberg Scholar, and Claire Ralph, Lecturer in Computing and Mathematical Sciences; Director, Career Development Center, on developing algorithms that address questions of explainability and algorithms that affect social change. [Caltech story]
IRCA Best Paper Awards
Two teams of Caltech researchers have won three International Conference on Robotics and Automation (ICRA) Best Paper Awards in multiple categories along with the overall best paper award. The ICRA is the largest and most prestigious robotics conference of the year. Awards are given on the basis of technical merit, originality, potential impact on the field, clarity of the written paper, and quality of the presentation. Maegan Tucker, Ellen Novoseller, Claudia Kann, Yanan Sui, Yisong Yue, Joel Burdick, and Aaron Ames, have won the ICRA Best Conference Paper Award and the ICRA Best Paper Award on Human-Robot Interaction (HRI) for their paper entitled "Preference-Based Learning for Exoskeleton Gait Optimization." Amanda Bouman, Paul Nadan, Matthew Anderson, Daniel Pastor, Jacob Izraelevitz, Joel Burdick, and Brett Kennedy, have won the ICRA Best Paper Award on Unmanned Aerial Vehicles for their paper entitled "Design and Autonomous Stabilization of a Ballistically Launched Multirotor." [Virtual Award Ceremony]
"Neural Lander" Uses AI to Land Drones Smoothly
Professors Chung, Anandkumar, and Yue have teamed up to develop a system that uses a deep neural network to help autonomous drones "learn" how to land more safely and quickly, while gobbling up less power. The system they have created, dubbed the "Neural Lander," is a learning-based controller that tracks the position and speed of the drone, and modifies its landing trajectory and rotor speed accordingly to achieve the smoothest possible landing. The new system could prove crucial to projects currently under development at CAST, including an autonomous medical transport that could land in difficult-to-reach locations (such as a gridlocked traffic). "The importance of being able to land swiftly and smoothly when transporting an injured individual cannot be overstated," says Professor Gharib who is the director of CAST; and one of the lead researchers of the air ambulance project. [Caltech story]
Creating a "Virtual Seismologist"
Professor Yisong Yue is collaborating with Caltech seismologists to use artificial intelligence (AI) to improve the automated processes that identify earthquake waves and assess the strength, speed, and direction of shaking in real time. Professor Yue explains, “the reasons why AI can be a good tool have to do with scale and complexity coupled with an abundant amount of data. Earthquake monitoring systems generate massive data sets that need to be processed in order to provide useful information to scientists. AI can do that faster and more accurately than humans can, and even find patterns that would otherwise escape the human eye.” [Read the full Q&A]
Meet the 2017 Amazon Fellows
Four graduate students from the Computing and Mathematical Sciences (CMS) Department and one from the Electrical Engineering (EE) Department have been selected as 2017 Amazon Fellows. This fellows program is the result of a partnership between Caltech and Amazon AWS around Machine Learning and Artificial Intelligence. The EE fellow is Srikanth Tenneti who is exploring the potential of deep learning for Direction of Arrival applications, and extending Ramanujan Sums based techniques for multi-dimensional periodicity extraction. CMS graduate student Navid Azizan Ruhi is researching faster optimization algorithms for machine learning. He is looking forward to visiting Amazon AI as a fellow and exchanging ideas with their researchers. Computer science graduate student Hoang Le is developing methods for efficient and intelligent sequential decision making in realistic systems. Florian Schaefer, whose focus is applied and computational mathematics, is researching the interface of statistical estimation and the design of fast algorithms. Control and dynamical systems graduate student Ellen Feldman, working with Professor Joel Burdick, has used part of the funding to present her research at the Society for Neuroscience annual meeting and looking forward to other future opportunities to share her research.
P. P. Vaidyanathan
Navid Azizan Ruhi