Prof. Sara Bernardini, Royal Holloway University of London
Large-Scale Autonomous Search-and-Tracking via AI Planning andProbabilistic Reasoning
Search-And-Tracking (S&T) is the problem of searching for a mobile target and tracking it to its destination once it is found. Since S&T platforms face many sources of uncertainty and operational constraints,progress in the field has been restricted to simple and unrealistic scenarios. In this talk, we propose a new hybrid approach to S&T that allows us to successfully address large-scale and complex S&T missions. The probabilistic structure of S&T is compiled into a deterministic planning model, and Bayesian inference is directly incorporated in the planning mechanism. Thanks to this tight integration between AI planning and probabilistic reasoning, we are able to exploit the power of both approaches. AI Planning provides the tools to efficiently explore big search spaces, while Bayesian inference, by readily combining prior knowledge with observable data, allows the planner to make more informed and effective decisions. We offer experimental evidence of the power of our approach by comparing it against other deterministic approaches (CP) and purely probabilistic methods (POMDPs). We also show how our technique can be easily extended to a multi-agent scenario.
Sara is a Senior Lecturer in Artificial Intelligence in the Department of Computer Science of Royal Holloway University of London and a Visiting Faculty member in the Computer Science and Artificial Intelligence Laboratory (CSAIL) of Massachusetts Institute of Technology (MIT). Sara's research interests focus on designing and engineering intelligent and autonomous systems for real-world applications and lie at the intersection between different areas: autonomous decision-making, cognitive robotics and intelligent control. Sara has significant experience in developing technologies for unmanned operations in extreme environments and in collaboration with industry and end users. As a research scientist at NASA Ames Research Center, she concentrated on equipping planetary rovers with high-level intelligent behaviour by integrating mission planning plan execution monitoring. At King's College London and Royal Holloway, she has specialized in autonomous surveillance missions using mobile robots, both in the aerial and the underwater domains, and worked in close collaboration companies such as BAE Systems and SeeByte. She is the Principal Investigator of the project SHARPA: Shared Autonomy via Robust Task Planning and Argumentation, funded by EPSRC. She has published more than forty peer-reviewed papers on autonomy in premier journals such as "Artificial Intelligence" and top conferences such as the "AAAI Conference on Artificial Intelligence?".