Rachid Alami
My topics of interest and my contributions, in collaboration with my colleagues and students, focus essentially on cognitive robotics (AI-enabled robotics): control architectures for autonomy, models, schemes and algorithms for robot decisional autonomy in various contexts and applications ranging from the remote autonomous robot, to multi-robot systems, to robots collaborating with humans.
Decisional abilities for Cognitive and Interactive Robots which share the task, the decision and the space with humans (Decision for HRI)
The scientific challenge here is to devise and build the cognitive and interactive abilities to allow pertinent, transparent, verifiable and legible behaviors for a robot that is able to perform collaborative tasks with humans. My contributions to Human-Robot Interaction and Collaboration cover several topics:
- Models, Architectures and High-level Control for Human-Robot Joint Action
- Human-Aware / Socially-Aware Motion, Manipulation and Task Planning for robots in presence and/or in synergy with humans
- Theory of Mind, Perspective-Taking and Affordances
- User studies applied to Human-Robot Joint Action
Combined Task and Motion Planning
I have contributed in various ways to this very challenging problem which involves Intricate Symbolic and Geometric Reasoning.
- A Formulation of the Manipulation Planning Problem and very early implementations of CTAMP planners
- The Asymov Planner: The first planner wich integrated effectively in unique planning framework, and in a principled manner, Symbolic and Geometric planning based on the formulation of the Manipulation Planning Problem.
- More recent contributions to Combined Task and Motion based on a refined reasoning of side effects of robot manipulation actions.
- Specific CTAMP based Task Planners: for the Handover task, for the transfer by Pivoting of heavy objects.
Control Architectures and tools for building autonomous robot systems
I have contributed in various ways, at the conceptual and practical level, to the elaboration of various Control Architectures for autonomous robots:
- Robot Control Architectures for cognitive robots: Initial and generic multi-level architectures integrating planning and time-bounded decision modules for autonomous robots.
- Robot Control Architectures for HRI: Specific contributions to the control of Cognitive and Interactive Robots that achieve tasks in presence and /or in collaboration with humans.
- Control Architectures for Multi-robot Systems: control architectures for Distributed Cooperative Multi-robot Systems.
- Control Architectures for Remote Autonomous Robots: Specific contributions to the control of Remote Autonomous Robots including Mars Rovers.
- Tools for Robotic Software Architectures: A modular design approach and associated tools for building Software Architectures for Autonomous Robots.
- Programming and Control of Flexible Assembly Cells: Contributions to programming and control of Flexible Assembly Cells composed of several robots, sensors and auxiliary devices; a Hierarchical control architecture allowing online decision and refinement of actions.
Task and Motion Planners for various contexts
- Action and Task Planning: HTN planning for multi-robot systems and human-robot interaction, Contribution to task planning algorithms taking into account uncertainty .
- Motion Planning: Kinodynamic motion planning for UAVs, Planning Mobile Robot Navigation, Planning Grasps, Pick&Place and Assembly Tasks.