Semantic search
| PrjDescription | PrjStudMax ?prjResArea | PrjTutor | PrjStudMin | |
|---|---|---|---|---|
| Exploratory data analysis by genetic feature extraction | Understanding the waves in EEG signals is an hard task and psicologists often need automatic tools to perform this task. In this project we are interested in using a genetic algorithm developed for P300 feature extraction in order to extract useful informations from Error Potentials. The project is a collaboration with the psicology department od Padua University. *B. Dal Seno, M. Matteucci, and L. Mainardi. "A genetic algorithm for automatic feature extraction in P300 detection" (http://home.dei.polimi.it/dalseno/papers/2008/ijcnn08.pdf) *B. Dal Seno, M. Matteucci, L. Mainardi, F. Piccione, and S. Silvoni. "Single-trial P300 detection in healthy and ALS subjects by means of a genetic algorithm" (http://home.dei.polimi.it/dalseno/papers/2008/grazGa08.pdf) | MatteoMatteucci | 1 | |
| Extended Kalman Filtering on Manifolds | Extended Kalman filtering is a well known technique for the estimation of the state of a dynamical system also used in robotics for localization and mapping. However in the basic formulation it assumes all variables to live in an Euclidean space while some components may span over the non-Euclidean 2D or 3D rotation group SO(2) or SO(3). It is thus possible to write tha Extended Kalman filter to operate on Lie Groups to take into account the presence of manifolds (http://www.ethaneade.org/latex2html/lie/lie.html). We are interestend in investigation this further applying it to the EKF-SLAM framework we have developed. '''''Material:''''' *papers about Manifold based optimization and space representations *C++ framework for EKF-SLAM '''''Expected outcome:''''' *An extended Kalman filter which uses this new representation '''''Required skills or skills to be acquired:''''' *Good mathematical background *C++ programming under Linux | MatteoMatteucci SimoneCeriani DavideCucci |
1 | |
| Extraction | This project is aimed at removing the ontology building bottleneck, semi-automatically building an ontology starting from a set of textual documents related to a specific domain. | DavideEynard MatteoMatteucci |
||
| Gestures in Videogames | Analysis of gestures and facial expressions of people involved in playing a videogame (TORCS) | MatteoMatteucci MaurizioGarbarino |
||
| HRVCar | MatteoMatteucci | |||
| HeadsetControlForWheelChair | Study how it would be possible to control a wheelchair with a low-cost headset detecting biosignals such as NIA | MatteoMatteucci | ||
| I.DRIVE Data Logger | This project concerns the logging of data coming from multiple sensor installed in a car. This data are used to analyze the behaviours and the reactions of the driver | MatteoMatteucci | ||
| Indoor localization system based on a gyro and visual passive markers | This project aims at developing autonomous moving systems based on a gyro and passive markers | MatteoMatteucci SimoneCeriani DavideMigliore |
||
| Information geometry and machine learning | In machine learning, we often introduce probabilistic models to handle uncertainty in the data, and most of the times due to the computational cost, we end up selecting (a priori, or even at run time) a subset of all possible statistical models for the variables that appear in the problem. From a geometrical point of view, we work with a subset (of points) of all possible statistical models, and the choice of the fittest model in out subset can be interpreted as a the point (distribution) minimizing some distance or divergence function w.r.t. the true distribution from which the observed data are sampled. From this perspective, for instance, estimation procedures can be considered as projections on the statistical model and other statistical properties of the model can be understood in geometrical terms. Information Geometry (1,2) can be described as the study of statistical properties of families of probability distributions, i.e., statistical models, by means of differential and Riemannian geometry. Information Geometry has been recently applied in different fields, both to provide a geometrical interpretation of existing algorithms, and more recently, in some contexts, to propose new techniques to generalize or improve existing approaches. Once the student is familiar with the theory of Information Geometry, the aim of the project is to apply these notions to existing machine learning algorithms. Possible ideas are the study of a particular model from the point of view of Information Geometry, for example as Hidden Markov Models, Dynamic Bayesian Networks, or Gaussian Processes, to understand if Information Geometry can give useful insights with such models. Other possible direction of research include the use of notions and ideas from Information Geometry, such as the mixed parametrization based on natural and expectation parameters (3) and/or families of divergence functions (2), in order to study model selection from a geometric perspective. For example by exploiting projections and other geometric quantities with "statistical meaning" in a statistical manifold in order to chose/build the model to use for inference purposes. Since the project has a theoretical flavor, mathematical inclined students are encouraged to apply. The project requires some extra effort in order to build and consolidate some background in math, partially in differential geometry, and especially in probability and statistics. Bibliography # Shun-ichi Amari, Hiroshi Nagaoka, Methods of Information Geometry, 2000 # Shun-ichi Amari, Information geometry of its applications: Convex function and dually flat manifold, Emerging Trends in Visual Computing (Frank Nielsen, ed.), Lecture Notes in Computer Science, vol. 5416, Springer, 2009, pp. 75–102 # Shun-ichi Amari, Information geometry on hierarchy of probability distributions, IEEE Transactions on Information Theory 47 (2001), no. 5, 1701–1711. | MatteoMatteucci LuigiMalago |
1 | |
| Integration of scanSLAM and ARToolKit in the MoonSLAM framework | The project aims at integrating the scanSLAM algorithm with the ARToolKit localization system in the MoonSLAM framework for multi sensor SLAM. | MatteoMatteucci SimoneCeriani |
||
| Interpretation of facial expressions and movements of the head | MatteoMatteucci | |||
| LARS and LASSO in non Euclidean Spaces | LASSO (1) and more recently LARS (2) are two algorithms proposed for linear regression tasks. In particular LASSO solves a least-squares (quadratic) optimization problem with a constrain that limits the sum of the absolute value of the coefficients of the regression, while LARS can be considered as a generalization of LASSO, that provides a more computational efficient way to obtain the solution of the regression problem simultaneously for all values of the constraint introduced by LASSO. One of the common hypothesis in regression analysis is that the noise introduced in order to model the linear relationship between regressors and dependent variable has a Gaussian distribution. A generalization of this hypothesis leads to a more general framework, where the geometry of the regression task is no more Euclidean. In this context different estimation criteria, such as maximum likelihood estimation and other canonical divergence functions do not coincide anymore. The target of the project is to compare the different solutions associated to different criteria, for example in terms of robustness, and propose generalization of LASSO and LARS in non Euclidean contexts. The project will focus on the understanding of the problem and on the implementation of different algorithms, so (C/C++ or Matlab or R) coding will be required. Since the project has also a theoretical flavor, mathematical inclined students are encouraged to apply. The project may require some extra effort in order to build and consolidate some background in math, especially in probability and statistics. Picture taken from (2) Bibliography # Tibshirani, R. (1996), Regression shrinkage and selection via the lasso. J. Royal. Statist. Soc B., Vol. 58, No. 1, pages 267-288 # Bradley Efron, Trevor Hastie, Iain Johnstone and Robert Tibshirani, Least Angle Regression, 2003 | MatteoMatteucci LuigiMalago |
1 | |
| Low-cost IMU | Characterization of a low-cost IMU module developed by AIRLab; use of such module to extract high-level information about human behavior. | MartinoMigliavacca MatteoMatteucci |
||
| Machine Learning for Crop Weed Classification | Automatic detection of crops and weeds by using image data | MatteoMatteucci | ||
| MoonSLAM Reengineering | In the last three years a general framework for the implementation of EKF-SLAM algorithm has been developed at the AIRLab. After several improvements it is now time to redesign it based on the experience cumulated. The goal is to have an international reference framework for the development of EKF based SLAM algorithms with multiple sensors (e.g., lasers, odometers, inertial measurement ) and different motion models (e.g., free 6DoF motion, planar motion, ackerman kinematic, and do on). The basic idea is to implement it by using C++ templates, numerically stable techniques for Kalman filtering and investigation the use of automatic differentiation. It should be possible to seamlessly exchange motion model and sensor model without having to write code beside the motion model and the measurement equation. '''Material''' *lots of theoretical background and material *an existing (and working) C++ implementation of the framework '''Expected outcome:''' *a C++ library for the implementation of generic EKF-SLAM algorithms '''Required skills:''' *Experienced C++ programming under Linux | MatteoMatteucci | 1 | |
| Multimodal GUI for driving an autonomous wheelchair | This project pulls together different Airlab projects with the aim to drive an autonomous wheelchair (LURCH - The autonomous wheelchair) with a multi modal interface (Speech Recognition, Brain-Computer Interface, etc.), through the development of key software modules. The work will be validated with live experiments. *R. Blatt et al. ''Brain Control of a Smart Wheelchair'' (http://tinyurl.com/ygldwun) | MatteoMatteucci SimoneCeriani DavideMigliore |
1 | |
| Object Recognition with Deep Boltzmann Machines | Deep Boltzmann machines for classification tasks | MatteoMatteucci FrancescoVisin |
||
| Odometric system for robots based on laser mice | We developed an odometric system for robots by combining the reading of several laser mice. The system consists of a master PIC-based board and several slave boards where the sensors employed in optical mice are located. The readings are collected on the PIC and sent on the serial port to a PC which elaborates and combines the x and y readings in order to obtain a x,y,theta estimation of the movement of the robot. The aim of the project is first to improve the current design of the PIC-based board, and realize a new working prototype, and then to implement and evaluate different algorithms able to estimate more precisely the x,y and theta odometric data from the mice readings. Experience with PIC-based systems and some experience with electronics circuits is a plus. Students are supposed to redesign the electronic board, improve the firmware of the PIC, and work on the algorithm that estimates the robot position on the PC. It would be also interesting to evaluate the possibility to embed the optimization and estimation algorithms in the firmware of the PIC in order to produce a stand-alone device. Ask the tutors of the project for extra material, such as data-sheets and other documentation. | MatteoMatteucci LuigiMalago MarcelloRestelli |
1 | |
| P300 BCI | Recovery, integration and adaptation of P300 BCI (hardware and software) stubs to generate a working interface for a speller. The aim is to develop a working prototype for an ALS affected patient. | MatteoMatteucci | 1 | |
| Poit cloud SLAM with Microsoft Kinect | Simultaneous Localization and Mapping (SLAM) is one of the basic functionalities required from an autonomous robot. In the past we have developed a framework for building SLAM algorithm based on the use of the Extended Kalman Filter and vision sensors. A recently available vision sensor which has tremendous potential for autonomous robots is the Microsoft Kinect RGB-D sensor. The thesis aims at the integration of the Kinect sensor in the framework developed for the development of a point cloud base system for SLAM. '''Material:''' *Kinect sensor and libraries *A framework for multisensor SLAM *PCL2.0 library for dealing with point clouds '''Expected outcome:''' *Algorithm able to build 3D point cloud representation of the observed scene *Point clouds processing could be used to improve the accuracy of the filter as well '''Required skills or skills to be acquired:''' *Basic background in computer vision *Basic background in Kalman filtering *C++ programming under Linux | MatteoMatteucci | 1 |
