Biography
Lydia E. Kavraki is the Kenneth and Audrey Kennedy Professor of Computing and professor of Computer Science, Electrical & Computer Engineering, Mechanical Engineering, and Bioengineering. She also serves as the director of the Ken Kennedy Institute at Rice University. Kavraki received her B.A. in Computer Science from the University of Crete in Greece and her Ph.D. in Computer Science from Stanford University, working with Professor Jean-Claude Latombe. She was a research associate at Stanford University before moving to Rice.
Kavraki works broadly in robotics, computational biomedicine, and physical AI. She has authored more than 300 peer-reviewed journal and conference publications and is one of the authors of the widely used robotics textbook “Principles of Robot Motion” published by MIT Press. Work in her group has produced the Open Motion Planning Library (OMPL), an open-source library of motion planning algorithms used worldwide. Besides this library her group maintains several web servers for computational biomedicine applications. Kavraki’s research has been funded by NSF, NIH, ARO, DOD, NASA, industry, and the Cancer Prevention and Research Institute of Texas (CPRIT). Kavraki's more than 40 postdocs and PhD students have gone on to faculty positions at top universities, industry research labs, startups, and large software companies. Through her role at the Ken Kennedy Institute, she brings faculty together across seven schools and twenty-seven departments to shape large research projects and future directions in AI, data, and computing at Rice University.
Kavraki is a member of the American Academy of Arts and Sciences, the National Academy of Medicine (NAM), Academia Europaea, and the Academy of Athens. She has served the Academies in multiple roles, including serving as a vice-chair and chair of Section 1 of NAM, founding member of the Health and Technology Interest Group of NAM, and member of the Board of Mathematical Sciences and Analytics.
Kavraki received the IEEE Frances E. Allen medal in 2023. She received the Association for Computing Machinery (ACM) Grace Murray Hopper Award (2000), the ACM Athena Lecturer Award (2017), and the ACM/AAAI Allen Newell Award (2020). Kavraki received the Early Academic Career Award from the IEEE Society on Robotics and Automation (2002) and Robotics Pioneer Award from the IEEE Robotics and Automation Society (2020). Other awards include an NSF CAREER award, a Sloan Fellowship, a Whitaker Investigator Award, recognition as a top TR100 investigator from the MIT Technology Review Magazine, recognition as a Brilliant 10 Scientist from the Popular Science Magazine, and the Anita Borg ABIE Technical Leadership Award. At Rice University, she is the recipient of the Charles Duncan Award for Excellence in Research and Teaching, the Presidential Mentorship Award, the Outstanding Faculty Research Award from the Engineering School, and the university-wide Faculty Award for Excellence in Research, Teaching and Service (2022). She has been recognized in Houston with BioHouston’s Women in Science Award. Kavraki is a Fellow of ACM, a Fellow of IEEE, a Fellow of the American Association for the Advancement of Science (AAAS), a Fellow of the Association for the Advancement of Artificial Intelligence (AAAI), and a Fellow of the American Institute for Medical and Biological Engineering (AIMBE).
More information about Kavraki's research can be found at the Kavraki Lab website.
Research
In robotics, Kavraki is interested in enabling robots to work with people and in support of people. Her research develops the underlying methodologies for achieving this goal: algorithms for motion planning for high-dimensional systems with kinematic and dynamic constraints, novel methods for reasoning under sensing and control uncertainty, integrated frameworks for long-horizon planning, methods for learning and for using experiences, and ways to instruct robots at a high level and collaborate with them. With her work, she seeks to develop the science of autonomy at all levels, as she believes that robots should seamlessly move between different degrees of autonomy depending on their interaction with humans and the task at hand. Starting with the Probabilistic Roadmap Planner, which solved motion planning problems for robotic manipulators in minutes, Kavraki’s seminal work on sampling-based methods for robot motion planning has dominated the field for two decades and is credited with bringing planning times down to seconds. In an engineering feat in 2024, her group produced ultra-fast planning methods that, for the first time, brought planning times for complex manipulations to microseconds on conventional processors. This work opens exciting new opportunities in making robots able to work safely alongside humans. Kavraki studies the incorporation of such planners in neuro-symbolic and learning frameworks. In the past, Kavraki has worked in multi-robot planning, assembly planning, manufacturing, industrial automation, and parts handling with conventional and non-conventional methods, such as microelectromechanical systems. She was involved in one of the first robotic systems for stereotactic radiosurgery, leading to technologies such as the Gamma Knife. Current applications in her lab include enabling robots to work alongside nurse instructors for nurse training and alongside astronauts in space habitats (in collaboration with NASA). Kavraki’s group has produced and now maintains several open-source projects: OMPL, a library that supported sampling-based motion planning; PlannerArena, a tool for benchmarking sampling-based planners; Robowflex, a library that makes using MoveIt for motion planning easy; MotionBenchmaker, an extensible, easy-to-use tool to generate and benchmark datasets for manipulation problems; HyperPlan, hyperparameter optimization framework for selecting, tuning, and optimizing motion planning performance; and TMKit for task and motion planning. All of the above are heavily used in industry and academia.
In biomedicine, Kavraki uses a robotics-engineering-inspired approach to develop computational methods and tools to model protein structure and function, understand biomolecular interactions, aid the process of medicinal drug discovery, analyze the molecular machinery of the cell, and help integrate biological and biomedical data for improving human health. Hidden in her work is modeling molecular entities with robotics methods, adding a filtering layer of geometry, and speeding up calculations. Her protein function annotation Labelhash webserver has been operating for 15 years, while the recent docking protocols APE-Gen and DINC are highly successful for large ligands and peptidomimetics. Recently, her group has studied protein function through the lens of the proteins involved in the immune system. Her work, in collaboration with researchers at the Texas Medical Center, has applications, among others, in personalized immunotherapy, vaccine design, and the discovery of novel therapeutics for viral infections and asthma. Kavraki is also interested in the prediction of molecular structure from mass spectra alone and the prediction of drug metabolites. This latter interest stemmed from her earlier work on metabolic pathway discovery. Kavraki’s group maintains the APE-Gen 2.0 webserver for docking peptide ligands to HLAs, the DINC webserver for docking large ligands, the DINC-COVID webserver for ensemble docking to SARS-CoV-2 proteins, and the Label Hash webserver for protein function annotation. The group also distributes EnGens, a computational framework for the generation and analysis of representative protein conformational ensembles; HLA-Arena, a customizable environment for the structural modeling and analysis of peptide-HLA complexes for cancer immunotherapy; and SARS-Arena for the selection and structural HLA modeling of conserved peptides from SARS-related coronaviruses for novel vaccine development.
Through the confluence of algorithms, statistical reasoning, formal methods, machine learning, data science, and, importantly, physics modeling, Kavraki and her collaborators seek to understand how computers can reason effectively and robustly about problems in the real world. Kavraki is particularly interested in the development of physical AI, that is AI that considers the constraints of the physical world. Physical AI will be pivotal for future research in robotics and biomedicine.
Participating in the discussion about the risks of AI, Kavraki explores the social and ethical implications of her research, and the often-unintended consequences of work in AI and machine learning. Recently, her group examined privacy concerns for robots that employ their sensors while navigating environments where humans are present. In another work, the group documented the biases in the training sets of neural predictors in immunotherapy, drawing attention to the potential therapeutic consequences of this bias to specific populations.
More information about Kavraki's research can be found at the Kavraki Lab website.
Industry Impact
The impact of Kavraki’s work in robotics is hard to measure as her methods are ubiquitous, are now included in textbooks, and are often mentioned without a reference. Her group has built the Open Motion Planning Library (OMPL) and has maintained this library since 2008. OMPL has an estimated 3,000 users in academia and industry (the estimate is conservative and is based on the users of the online manual of OMPL). The number of library downloads is impossible to obtain, as OMPL is part of ROS (the Robot Operating System), and many users get the library through the ROS package manager. ROS runs in practically every modern robot. Students, researchers, and companies have used OMPL, which is distributed with a permissive BSD license, in an estimated 30 robotic systems. OMPL has been integrated with popular robotics software and simulators, including MoveIt, OpenRAVE, CoppeliaSim, MORSE, the Kautham Project, VEROSIM, AIKIDO, EXOTica, the Robotics Library, and others. As OMPL has gained traction and is used under the hood in several products (including commercial products such as MoveItPro), companies are selling consulting services for OMPL. OMPL has had over 20 external contributors and is now considered a robotics-community project.
Kavraki has worked with NASA to develop and apply sampling-based algorithms to humanoid robots developed for space operations. Of special mention is Robonaut2, NASA’s robotic assistant at the International Space Station.
In the computational biomedicine domain, Kavraki’s APE-Gen docking tool is the leader in the computation of protein-MHC interactions as it can handle post-translational modifications. The tool is used at MD Anderson to predict binding peptides for downstream analysis in personalized immunotherapy pipelines.
The Ken Kennedy Institute
Kavraki provides technical leadership at Rice University through her position as the director of the Ken Kennedy Institute. The Ken Kennedy Institute, established in 1986, is dedicated to solving critical global challenges with collaborative approaches that focus on research innovation in AI, Data, and Computing. The Ken Kennedy Institute is the virtual home of over 250 faculty members and senior researchers at Rice University, spanning engineering, computer science, mathematics, statistics, natural sciences, humanities, social sciences, business, architecture, and music. The Institute brings together faculty from 7 schools and 27 departments at Rice, and it catalyzes research collaborations across the conventional boundaries of university, department, center, and laboratory. It coordinates large research proposals and sustains an eco-system that facilitates research endeavors. The Institute enables new conversations through its Distinguished Lecture Series, technical workshops, and two annual conferences: the Ken Kennedy Institute's AI in Health Conference and the Energy HPC Conference. The Institute aids workforce development through its summer Machine Learning Boot Camp. It works closely with industry to bring promising ideas to market and develop academic, industry, and community partnerships. Lastly, the Ken Kennedy Institute is an advisor to the Office of Research Computing for securing the computational resources that allow Rice researchers to achieve preeminence.
Service, Editorial Positions, and Mentoring
Kavraki is currently an elected member of the IEEE Robotics and Automation Administrative Committee, which manages the IEEE Robotics and Automation Society. She participates in the Finance Board, the Publications Board, and the Industry Board. Kavraki has served IEEE and ACM in various roles, including selection committees for various grades and awards of the societies.
Since 2019, she has been a member of the Board on Mathematical Sciences and Analytics of the National Academies (BMSA). In that role, she has been a reviewer of several projects. Kavraki was a member of the Committee on Emerging Science, Technology, and Innovation in Health and Medicine (CESTI) from 2019 to 2022. She is the founding chair of the Interest Group on Health and Technology of the National Academy of Medicine (2018-2022), and in that role, she organized several workshops: Data Powered Health (2019), Telemedicine and Telehealth: Accelerated Change in the Era of the Pandemic (2020) and Human Health and Equity in an Age of Robotics and Intelligent Machines (2021). She now serves as the chair of the Education of the Health Care & Science Workforce (2024-2026). Kavraki was a co-chair of the 2023 Annual Meeting of the Texas Academy of Science, Medicine, and Engineering.
Kavraki currently serves on the Board of Reviewing Editors for PNAS-Nexus of The National Academy of Sciences of the United States of America. She is Senior Associate Editor for Annual Reviews for Robotics, Control, and Autonomous Systems from Annual Reviews (Editor from 2015-2020); Associate Editor for Science Robotics from Science Journals; and on the Editorial Board for Springer Tracts in Advanced Robotics (STAR), from Springer – Verlag. She was an associate editor of the International Journal of Robotics Research from 2009 to 2022. Kavraki has been the program chair and the general chair of “Robotics: Science and Systems,” the premier robotics conference, and she has served on the program committee of numerous conferences in robotics, including ICRA, IROS, RSS, WAFR, ACC, and others. Currently, she is on the advisory board of RSS and WAFR. Kavraki has generously offered her services on review panels for proposals and award committees. She is also a member of the external advisory committee of two AI Institutes, one major research hospital, and several departments.
On the mentoring side, Kavraki has been a mentor for several faculty at all stages of their career through formal programs such as the mentoring program of the Gulf Coast Consortia and NSF ADVANCE.
Kavraki leads the NIH/NLM T 15 Training Program in Biomedical Informatics and Data Science under the auspices of the Keck Center of the Gulf Coast Consortia in Houston. This training grant involves nine undergraduate and five postdoctoral students.