ULB

Home
Curriculum Vitae
Publications
Research Interests
Projects and Awards
Teaching Activity
Thesis Proposals

Emanuele Garone

 

My research activity focused mostly on the following fields :
1) Model Predictive Control
2) Fault Tolerant Control Allocation
3) Remote Control and Estimation through unreliable networks
4) Cooperative path planning for heterogeneous vehicles

 

Model Predictive Control: The Model Predictive Control philosophy consists in computing, on the basis of actual state plant measurements, the “best” sequence of control moves such that plant forecasts fulfil specific performance and/or constraints requirements. The obtained control strategy is applied to the system until new plant measurements are available and then a new set of virtual input moves is computed. My research activity in this field has been mostly devoted to the development of MPC schemes for Linear Parameter Varying (LPV) systems subject to saturation constraints. Further directions of research includes the use of Sum of Squares for nonlinear MPC schemes, reference management of power networks, sensorless reference management for distributed agents and MPC schemes for switching systems.

 

Fault Tolerant Control Allocation: Actuator redundancy is an important issue to deal with in many real plants. For example, it is a very common matter in vehicle applications. A well-established way to handle overactuated systems consists of using a control law that specifies only the total control effort that has to be produced and separately solving the so-called Control Allocation Problem (CAP) i.e. the one of optimally distributing the desired total control effort over the available actuators. The main idea of this research line is to exploit the presence of redundant actuators is exploited to develop effective fault-tolerant reconfigurable control strategies.

 

Remote Control and Estimation through unreliable networks: Unlike the classical control paradigm, where components, i.e. sensors, controllers and actuators, are often co-located and therefore connected via dedicated channels (e.g. serial cables), spatially distributed systems makes use of general purpose networks to communicate and exchange information. As a consequence the communication introduces uncertainty as data may get delayed or lost in the communication. My research activity in this field was mainly devoted to the problem of control through erasure channel (channel in which packets may be lost accordingly to a Bernoulli variable). Further research directions in this regard optimal sensor selection topics and security topics (detection of malicious attacks).

 

Cooperative path planning for heterogeneous vehicles: The complexity of many applications envisioned for future autonomous vehicle networks requires a broad range of capabilities for individual units which cannot be implemented on a single platform class. Rather, it may be necessary to coordinate several specialized units to attain complex objectives in a reliable, timely, and efficient fashion. While considerable progresses have been made on cooperative control of networks of homogeneous vehicles, heterogeneous networks are still relatively poorly understood and represent the main topic of this research line. Preliminary interesting results have been obtained for the mission planning of a relevant class of vehicles network, arising from the combination of (i) a slow autonomous surface carrier (typically a ship) with long range operational capabilities and (ii) a faster flying vehicle (typically an UAV) with a limited operative range.


Last updated 28/04/2010