Reason
Represent tasks, goals, and constraints through symbolic and hierarchical models.
Reason. Plan. Act.
An Interactive Workshop on Physical AI
18-22 October 2026
Connect abstract AI reasoning with reliable decision-making and execution in complex physical environments.
Led by Dr. Sarah Keren, Dr. David Dovrat, Guy Azran, and Almog Anschel from CLAIR Lab
From intelligence to action
Building AI systems that operate in the physical world
The workshop integrates symbolic reasoning, geometric planning, perception, and execution into coherent decision-making systems, with an emphasis on modern planning and learning techniques for operating under uncertainty.
Hands-on learning
Turn ideas into physical behavior
Participants will implement and experiment with core algorithms, gaining practical insight into the complete reasoning-to-action pipeline.
Plan
Combine task and motion planning with perception and decision-making under uncertainty.
Act
Test planning and learning methods in realistic simulation and on physical robots.
Workshop objectives
From abstract reasoning to reliable physical action
The workshop gives participants a practical understanding of how Physical AI systems connect high-level goals with perception, geometry, dynamics, uncertainty, and real-time execution. By the end of the workshop, participants should be able to:
Model physical tasks
Express goals, actions, constraints, and dependencies using symbolic and hierarchical task representations.
Connect task and motion planning
Combine discrete task decisions with continuous geometry, collision constraints, and feasible robot motion.
Reason under uncertainty
Account for imperfect sensing, changing environments, and execution failures when choosing and updating actions.
Implement and evaluate
Build core algorithms in interactive notebooks and evaluate them in simulation and on physical robotic platforms.
Workshop format
Interactive by design
Designed for undergraduate and graduate students, researchers, and practitioners interested in AI, robotics, autonomous systems, machine learning, and control.
- Interactive Jupyter notebooks
- No installation required
- Realistic simulation environments
- Hands-on work with real robots
- Planning and learning under uncertainty
Full program
Tentative Workshop Schedule
A five-part progression from Physical AI foundations to integrated experiments on robotic platforms. Exact dates and session times will be announced.
Meet the team
Workshop organizers
The workshop is led by members of the Collaborative AI and Robotics Lab (CLAIR) at the Technion.
Dr. Sarah Keren
Workshop Lead · Head of CLAIR Lab
Dr. David Dovrat
Organizer · CLAIR Lab
Guy Azran
Organizer · CLAIR Lab
Almog Anschel
Organizer · CLAIR Lab
Questions and resources
Contact the organizers
For questions about the program, participation, accessibility, or registration, contact the workshop organizer.
Join the workshop
Registration
Registration is now open. Complete the form to apply for participation.
Register now