From today I’m PhD in Automation and operation research. The title of my thesis is: “Non linear sampled-data control for mobile robotics”
Keywords: sampled-data control, nonlinear systems, robotics
The PhD thesis deals mainly with the conception and realisation of an autonomous mobile platform for the inspection of a Tokamak machine (Autonomous Rover for Video Inspection – ARVI).
The work has been developed in the framework of a “study contract” of the Italian National Agency for New technologies (ENEA) with the University of Rome la Sapienza.
Starting from the constraints imposed by the environment (the experimental Tokamak machine for fusion energy generation available at the ENEA laboratories in Frascati) a track like platform of suitable dimensions (28 cm width, 26cm depth and 7cm height) has been designed and realized at the Laboratory of Systems and Control of the “Dipartimento di Informatica Automatica e Gestionale della Sapienza”. Nonlinear sampled data planning procedures and control strategies has been designed and implemented on the platform as well.
The planning procedure has been designed on the basis of the exactly computable sampled equivalent model of the car-like non holonomic mathematical model of the platform subject to a preliminary state feedback. Asymptotic tracking of the planned state trajectories has been then obtained by multirate digital matching and stabilization techniques. Planning and control procedures are feed by measures from odometric and infrared sensors, accelerometers and visual signals for simultaneous localization and mapping of the platform on a web-page.
An augmented reality framework has been developed to manage the interaction between autonomous obstacle avoidance and informations from the external environment. Obstacle can be added and/or removed from the web-page monitor generating real-time updates of path planning and control procedures ensuring real and/or virtual obstacle avoidance.
The prototypal platform has been fully “in house” designed and realized: from CAD design to the construction of its parts, from the computation of the planning and control algorithms to their hardware implementation through ad-hoc measure and actuation devices.
A Patent request is in course of redaction.
In the line of developing monitoring and control software environments, as the augmented reality framework for ARVI, a simulation environment for flying control systems has been developed and applied to verify the performances of control systems for tracking aerial trajectories of a tri-rotor, ref.1, for maneuvering a PVTOL in presence of delays on the input, ref. 2, and for assuring sampled data regulation of elliptic orbits around the L2 translunar libration point, ref. 3.
Ref1. G. Gonnelli and R. Bonghi. A novel approach for aerial manipulation. In International CAE Conference, Pacengo del Garda (Verona), Italy, 2014.
Ref2. L. Ricciardi Celsi, R. Bonghi, S. Monaco, and D. Normand-Cyrot. On the exact steering of finite sampled nonlinear dynamics with input delays. In MICNON 2015 – Conference on Modelling, Identification and Control of Nonlinear Systems, volume 48, pages 674 – 679, Saint-Petersburg, Russia, 2015.
Ref3. L. Ricciardi Celsi, R. Bonghi, S. Monaco, and D. Normand-Cyrot. Sampled-data stabilization around the L2 translunar libration point. In 3rd IAA Conference on University Satellite Missions and CubeSat Workshop and International Workshop on Lean Satellite Standardization, Palazzo Rospigliosi, Roma, Italy, November 2015.
Copyright © 2016. All Rights Reserved.