Multi-agent Systems


In the field of multi-agent systems, we seek to design teams of robotic agents capable of solving a number of problems including environmental monitoring, agricultural tasks such as crop spraying, search and rescue in disaster response, law enforcement and surveillance. While the technology for the construction of unmanned vehicles is now relatively mature, endowing these machines with the intelligence necessary to complete the above tasks is an extremely hard problem, which has become one of the major challenges in the fields of control and artificial intelligence.

Our focus is on the design of algorithms for robotic agents with minimal sensing and communication capabilities, such as micro UAVs. Randomization is a key element in our approach to overcome such limitations.

Networked Control Systems

ncs In a NCS, sensors, actuators and controllers communicate with each other through a shared network. While reducing costs with cables and maintainance, this framework poses great performance challenges from the control point of view. Possible application areas include industrial plants, embedded systems and mobile robotics.

Our research on NCSs aims at cross-layer design strategies in which controller and network take advantage of the information about each other's statuses.

Stochastic Hybrid Systems

automaton In a SHS, continuous and discrete-valued variables interact and are subject to transitions driven by random phenomena. The most familiar example is the case of a computer (where variables are discrete) controlling a physical system (where variables are continuous).

We use SHSs as a modeling tool for various processes from control engineering to biology and use the framewok of Markov processes in general spaces for analysis and synthesis purposes.

Former and Current Students

Student ThesisStarting SemesterGraduation Semester
Wendy Herrera Filtering in Stochastic Hybrid Systems (PhD) 2013/12018/1
Petrus Abreu Design of Soft Sensors under Intermittent Excitation (MSc) 2014/12016/2
José Pácola Volatility and Risk Analysis of Low and High Grade Iron Ore (MSc) 2014/12016/2