05/11/2026
WHO: Dr. Ayal Taitler, Ben-Gurion University of the Negev
WHEN: Thursday, November 5th 2026 at 12:00
WHERE: BUILDING 503 (Computer Science) AUDITORIUM
Title: Dynamics-Aware Hybrid Planning for Robotic Missions
Abstract: Robots operate in the real world, which is inherently hybrid, combining discrete decisions with continuous motion and dynamics. Moreover, robotic missions often require reasoning over long time horizons while satisfying temporal constraints and spatial constraints that specify regions rather than exact states. These constraints couple decisions across different parts of the mission: where and when a robot enters a region affects both the motions that precede and follow it and their dynamic feasibility. Hybrid planning provides tools for reasoning over such long horizons, but usually relies on simplified models of motion in order to keep the planning problem tractable. On the other hand, accurate nonlinear dynamic models can be used for short-term motion planning and control, but incorporating them directly into long-term planning can make the problem computationally intractable.
In this talk, I will present our work toward dynamics-aware hybrid planning. The main idea is to combine different levels of model fidelity, using simple abstractions for efficient long-term reasoning and accurate nonlinear models where they are required for physical execution. By combining hybrid planning, optimal control, and learning, we aim to retain long continuous horizons, regional and temporal constraints, concurrency, and synchronization, while producing plans that more faithfully represent what the physical system can execute. The approach is demonstrated on simulated robotic missions and physical robot experiments.
Bio: Dr. Ayal Taitler is a faculty member in the Department of Industrial Engineering and Management at Ben-Gurion University of the Negev, where he leads the Decision-making, Robotics, and Intelligent Vehicles (DRIVE) Lab. His research focuses on decision-making in complex systems, spanning AI planning, reinforcement learning, robotics, control, and intelligent transportation systems, with particular interest in hybrid discrete-continuous and stochastic problems. Prior to joining Ben-Gurion University, Ayal was a Lyon Sachs Postdoctoral Fellow at the University of Toronto and a teaching fellow at the Technion's Faculty of Electrical and Computer Engineering. He received his Ph.D. in Autonomous Systems and Robotics from the Technion, and his Master's and Bachelor's degrees from the Technion's Faculty of Electrical and Computer Engineering.
