Caltech Young Investigators Lecture
Decentralized Multi-Robot Active Search and Rescue
Abstract: One of the main challenges of active search and rescue with multiple agents in unknown environments is impracticality of central coordination due to the difficulties of connectivity maintenance. In the first part of this talk, I will propose two novel algorithms for aerial robots that allow them to communicate and make data-collection decisions without a central coordinator. I will show how we strategically introduce the concept of sparsity into our active learning algorithms to improve search time without letting sparsity assumptions limit exploration. So far, existing search algorithms often prioritize the location accuracy of objects of interest while other practical issues such as the reliability of object detection as a function of distance and lines of sight remain largely ignored. In the second part of my talk, I will introduce a novel search algorithm that addresses these issues for multiple ground robots considering sensory information from monocular optical imagery and depth maps. I will demonstrate the viability of our algorithms using simulation results, theoretical analysis and a video illustration in a game environment we have created.
Bio: Ramina Ghods is a postdoctoral fellow at the Robotics Institute at Carnegie Mellon University working with Jeff Schneider. Ghods works on devising active learning and tactical decision-making algorithms for autonomous multi-agent search problems that arise in the field of robotics. Her research interests include inference and estimation, active learning, sparse signal recovery, Bayesian Optimization, and machine learning. She completed her PhD in 2019 at Cornell University with Christoph Studer developing novel inference, estimation and initialization algorithms for applications in machine learning, imaging and wireless communications. Ghods received her bachelor's degree in Electrical Engineering at Sharif University of Technology, Tehran, Iran. She is a 2020 EECS rising star and a recipient of the 2013-2014 Cornell fellowship and 2014 Cornell Jacobs Scholarship. Throughout the years, she has been involved in a variety of outreach activities by organizations such as SWE, Girls Who Code, EYH, and CATALYST.
This talk is part of the Caltech Young Investigators Lecture Series, sponsored by the Division of Engineering and Applied Science.
Contact: Caroline Murphy email@example.com