Luca Cavanini

Model predictive control for the reference regulation of current mode controlled DC-DC converters

The control of DC-DC converters for high performance applications usually relies on Current Mode Control (CMC). Compared to voltage mode control, it guarantees automatic over-current protection, and a better closed loop stability, together with improved transient response. This paper presents a Model Predictive Control (MPC) algorithm for the regulation of the voltage reference signal in CMC. The control of pre-compensated systems via MPC is common in several fields, such as automotive and aerospace, and power electronics is a perfect candidate to exploit this hierarchical structure as well. Indeed, controllers for power converters are often coded in the integrated circuits, and cannot be changed. Furthermore the possible multirate structure allows to exploit the performance of MPC less affecting the computational cost. The paper describes the design of an MPC regulator for a synchronous buck converter, when a primal CMC is already coded. Performance improvements of the proposed controller are reported.

SLAM-based autonomous wheelchair navigation system for AAL scenarios

Commercial electric-power wheelchairs have become much cheaper in the recent years, likewise the availability of reduced size sensors to an affordable price has made their integration easier in this kind of vehicle. This paper presents the development of a smart navigation system applied to an electric wheelchair. The developed work falls within the Ambient Assisted Living field (AAL), which includes all the technologies whose aim is to improve the quality of life of people in the home environment, especially for the elderly and physically impaired. In particular, the present work is focused on the development of technological support aimed at improving the daily life of that population segment who has motor difficulties, and is forced to use personal mobility support systems such as a wheelchair. This system is able to localize the wheelchair while it is moving in an indoor environment. The system exploits a low cost hardware and an integrated open source software, which permit a cheap integration with already available electric wheelchairs.

Microgrid sizing via profit maximization: A population based optimization approach

In this paper we present a computational intelligence approach to solve the optimal sizing problem of grid connected microgrid (MG) components. A simulation model has been built for the MG and comprises households, solar photovoltaic (PV) plants, wind turbines (WT) and energy storage (ES) systems. The goal is the maximization of the long term economic benefits for the community being served by the MG. We choose to optimize the net present value (NPV) of the whole investment in a cost benefits analysis (CBA) scenario. In particular, due to the complexity and the high-dimensionality of the problem, we solved it using population based optimization techniques. We tested four different algorithms in their basic form, i.e. artificial bee colonies, particle swarm optimization, genetic algorithm and gravitational search algorithm, comparing their performances. The effectiveness of the approach is tested in a case study where the optimal ratings for the PVs, WTs and ESs are determined using real weather and electrical demand data in the central east part of Italy.

Robust control of piezostage for nanoscale three-dimensional images acquisition

Piezoelectric drivers are widely used in nanoscale image acquisition systems. A source of performance degradation of these drivers is hysteresis, which introduces low rate noise and causes a slow continuous drift, decreasing the quality of scanned images. In this paper a sliding mode control policy, based on an estimator of the perturbation due to hysteresis, is applied to the control of a piezostage in a three dimensional images acquisition system, in order to improve control performances and final scanning results. The presented solution has been experimentally tested and compared with the Proportional-Integrative-Derivative (PID) controllers provided with the piezoelectric acquisition system, showing a noticeable improvement both in the piezostage behavior and in the images quality.

A fast model predictive control algorithm for linear parameter varying systems with right invertible input matrix

Nowadays, Linear Parameter Varying (LPV) Model Predictive Control (MPC) represents a consolidated approach to optimally regulate multivariable nonlinear systems imposing constraints on inputs and outputs. The crucial drawback, in particular in embedded LPV-MPC, is represented by the required computational effort. The Quadratic Programming (QP) problem solved by MPC changes at each iteration, and its reconstruction increases drastically the time required to compute the control law. This paper proposes an algorithm to reduce the LPV-MPC computational complexity for a particular class of linear systems. By exploiting a coordinate transformation, the QP reconstruction can be avoided. The effectiveness of the novel method is shown on a benchmark set of random MPC problems, as well as on a real-world case study regarding the control of an heat exchanger cell.

Development and experimental validation of a LQG control for a pre-compensated multi-axis piezosystem

Positioning Piezoelectric Actuators (PAs) are extremely versatile high precision investigative tools exploited in the field of micro-nanotechnology and they are also widely used in highly accurate industry production. PAs systems require appropriate controllers allowing to obtain fast and high-precision positioning performances but usually commercial systems provide integrated Proportional-Integral (PI) controllers providing standard stability performances also in the presence of uncertainty and disturbance. Those PI controllers usually are not tunable or adjustable on the considered plant, limiting controlled system performances with respect to nominal PAs features. In this paper the authors propose to improve the PI pre-compensated system performances through an external Linear Quadratic Gaussian (LQG) controller, in order to modify PIs reference signals behaviour by a feedback control law. The considered system is a micro-nano resolution triaxial piezoelectric driver controlled by a commercial amplifier providing a set of internal analog PI controllers. In this paper, the PI pre-compensated plant is identified in the form of a Multi-Input Multi-Output (MIMO) Linear Time-Invariant (LTI) model, due to the linear behaviour of the actuated axes. Effectiveness of the proposed method is validated by simulation tests and experiments on the real system, comparing the proposed approach with respect to the PI pre-compensated plant performances.

A model predictive control for a multi-axis piezo system: Development and experimental validation

This paper presents a Model Predictive Control (MPC) strategy for a triaxial piezoelectric actuators (PAs) system. PAs systems require appropriate controllers to guarantee fast and high-precision positioning performances avoiding effects of non-linearities. Typically, commercial systems provide integrated Proportional-Integral (PI) controllers guarantying to maintain system stability in the presence of uncertainty and disturbance. MPC owes its success to the ability of optimally regulate multivariable systems through the minimization of a Quadratic Programming (QP) problem subjected to prescribed constraints. Alternatively, unconstrained MPC eliminates constrains from the problem, reducing the number of operations elapsed to compute the solution. The aim of this work is to design an unconstrained MPC for a 3-DOF PA replacing PI controllers to improve control performances by a smaller increase of required computational effort. The system is described by a Multi-Input Multi-Output (MIMO) Linear Time-Invariant (LTI) model, experimentally identified by the open-loop real plant response. Effectiveness of the proposed method is validated by simulation tests and experiments on the real system, comparing MPC with PI controllers tuned to guarantee common PA stability requirements.

A QR-code localization system for mobile robots: Application to smart wheelchairs

A Smart Wheelchair (SW) is an electric powered wheelchair, equipped with sensors and computational capabilities, with the general aim of both enhancing independence and improving perceived quality of life of the impaired people using it. SWs belong to the class of semi-autonomous mobile robots, designed to carry the user from one location to another of his/her choice. For such systems, the localization aspect is of utmost concern, since GPS signal is not available indoors and alternative sensor sets are required. This paper proposes a low- cost artificial landmark-based localization system for mobile robots operating indoor. It is based on Quick Response (QR) codes, which contain the absolute position of the landmark: a vision system recognizes the codes, estimates the relative position of the robot (i.e., displacement and orientation) w.r.t. the codes and, finally, calculates the absolute position of the robot by exploiting the information contained in the codes. The system has been experimentally validated for self-localization of a smart wheelchair, and experimental results confirm that navigation is possible when considering an high QR-code density, while QR-code low density conditions permit to reset the cumulative odometry error.

rapros: A ROS Package for Rapid Prototyping

ROS framework lacks of an internal tool to design or test control algorithms and therefore developers have to test their algorithms on-line, directly on the robotic platform they are working with. This is not always safe and possible, and a rapid prototyping tool can help during the design phase. Users can develop their algorithms directly on the controller board and safely test them in a simulated scenario. Although some rapid prototyping tools exist in the ROS community, none of them take Simulink® into consideration. In this work the authors provide an open source Rapid Prototyping tool which integrates ROS and Simulink. The proposed package is useful for control designers, who are frequently used to exploit Simulink features for control deployment. The tool can be downloaded from https://github.com/gionatacimini/rapros.

A low cost mobile platform for educational robotic applications

This paper proposes a low cost mobile platform for robotic applications which is mainly addressed to educators and developers. Most of robotic models in the market use USB or serial standard protocol to retrieve sensor information, and this can generate delays on sensor readings and degradation of the overall performances. The proposed mobile robotic platform is based instead on the Controller Area Network, namely CAN-bus, which permits to obtain high performances and to increase flexibility and expandability for future technology integration.