Massimiliano Nolich

Algorithms and architectures for acoustic localization based on microphone array in service robotics

This paper describes an experimental mobile robot with acoustic source localization capabilities for surveillance and transportation tasks in indoor environments. The location of a speaking operator is detected via a microphone array based algorithm; localization information are passed to a navigation module which sets up a navigation mission using knowledge of the environment map. The system has been developed using a distributed architecture with TCP/IP message passing. We describe the hardware and software architectures, as well as the algorithms. Experimental results describing the system performance in localization tasks are reported.

Fast Genetic Scan Matching Using Corresponding Point Measurements in Mobile Robotics

In this paper we address the problem of aligning two partially overlapping surfaces represented by points obtained in subsequent 2D scans for mobile robot pose estimation. The measured points representation contains incomplete measurements. We solve this problem by minimizing an alignment error via a genetic algorithm. Moreover, we propose an alignment metric based on a look-up table built during the first scan. Experimental results related to the convergence of the proposed algorithm are reported. We compare our approach with other scan matching algorithms proposed in the literature, and we show that our approach is faster and more accurate than the others.

SPATIAL MAP BUILDING USING FAST TEXTURE ANALYSIS OF ROTATING SONAR SENSOR DATA FOR MOBILE ROBOTS

This paper presents a novel, fast algorithm for accurate detection of the shape of targets around a mobile robot using a single rotating sonar element. The rotating sonar yields an image built up by the reflections of an ultrasonic beam directed at different scan angles. The image is then interpreted with an image-understanding approach based on texture analysis. Several important tasks are performed in this way, such as noise removal, echo correction and restoration. All these processes are obtained by estimating and restoring the degree of texture continuity. Texture analysis, in fact, allows us to look at the image on a large scale thus giving the possibility to infer the overall behavior of the reflection process. The algorithm has been integrated in a mobile robot. However, the algorithm is not suitable for working during the mobile robot movement, rather it can be used during the period when the robot stays in a fixed position.

A Hard Real-Time Kernel for Motorola Microcontrollers

This paper describes a real-time kernel for running embedded applications on a recent family of Motorola microcontrollers. Both periodic and aperiodic real-time tasks are managed, as well as non real-time tasks. The kernel has been called Yartos, and uses a hard real-time scheduling algorithm based on an EDF approach for the periodic task; aperiodic tasks are executed with a Total Bandwith Server.

Effects of weather condition on aircraft emissions in climb phase

Aircraft trajectory optimization is highly sensitive to atmospheric conditions; pressure, relative humidity, temperature, wind intensity and direction have various influences on thrust needed and the resulting air pollutant emissions. The airline flight plans are generally pre-calculated before take off in order to optimize fuel consumption, using information from weather predictions that may not be accurate enough. In this paper an evaluation of weather prediction accuracy and, in the case of inaccurate predictions, a comparison of estimated emissions of some flights in climb phase for different weather conditions are presented. Weather data used are from National Oceanic and Atmospheric Administration (NOAA) public domain data, specifically the GRIB (Gridded Binary) files of the 20 Km RAP (Rapid Refresh) model, containing the analysis of real weather of a certain day/hour and the forecasts of the following 18 hours. In order to better understand the relation between weather conditions and aircraft emissions we report a comparison between estimated emissions (fuel, CO2 and NOX) of a real trajectory calculated with real weather data and with predicted weather data (forecasts for 1h, 3h and 6h). In order to evaluate accuracy of forecasts we consider radar reflectivity and wind. Regarding evaluation of the presence of potentially dangerous clouds (level 2 or more), a threshold filter has been used to select regions above 36 dBZ in the weather analysis and in a previous forecast related to the same hour. In the first step, the radar reflectivity and wind of real USA weather data from four days of June 2012 were compared with the forecasts, using the Tanimoto Similarity Index for measuring accuracy. Given the exact shape on a grid of the region (in this case, the current weather analysis) and its approximation (in this case the forecast), the Tanimoto Index (TI) is defined as the number of “pixels” of intersection on the number of pixels of the union of the two images. Then each one of the weather analyses for the 4 days considered (96 hours total) was compared with the forecasts for that time from 1 to 6 hours before, computing the Tanimoto index and the total cloud coverage with a threshold at 36 dBz. Furthermore, the wind intensity and direction forecasts were analyzed, and the mean value and variance of the difference between real weather condition and forecasts are presented. In a second step, from the analysis of the results of the first step, we selected some regions where cloud and wind analysis were substantially different from forecasts. In this scenario, the climb phases from real aircraft trajectories were collected from the FlightAware database. In the region of bad weather, we selected the trajectories that were significantly different from those made from the same aircraft in days of good weather. The related emissions were estimated and compared with the emissions of the same trajectory using forecasted weather. The emission estimation model is based on BADA (Base of Performance Data) from EUROCONTROL, ICAO and weather data.

Ergodic Continuous Hidden Markov Models for workload characterization

In this paper we present a novel approach for accurate characterization of the execution workload run by a computer. Usually, workload characterization is performed by measuring the type and amount of resources requested during a program execution (for instance the usage of CPU, I/O, network, etc.). The sequence of measures is then treated as a stochastic process and analyzed with statistical techniques. The novelty of our approach is that we instead use directly the sequence of memory references generated during the execution of a program. The sequences of memory references are treated as sequences of floating point numbers, and analyzed with signal processing techniques. In the feature extraction phase we use spectral analysis while in the pattern matching phase we use ergodic continuous hidden Markov models (ECHMMs). The ECHMM models estimated in an initial training phase can be used both for online workload classification of a running process and for synthetic traces generation. Several processes of the same workload are necessary to obtain an HMM model of the workload. The proposed algorithms is evaluated via trace driven simulations using the SPEC 2000 workloads. We show that ECHMMs describe address memory sequences; average classification accuracy is about 76% with eight different workloads.

Visual scene analysis using relaxation labeling and Embedded Hidden Markov Models for map-based robot navigation

A scheme for extracting environment features and performing their interpretation from visual data for mobile robot navigation is presented. Each frame of the low rate image stream acquired by the robot is processed as a separate image. Segmentation of the image is done using a graph-based approach in order to select the regions of interest (ROIs) of the visual scene. ROIs are processed to extract the edges of the objects using relaxation labeling. The obtained image is analyzed using a machine learning approach based on embedded HMMs. Experimental results are presented for an office environment.

A genetic-fuzzy algorithm for the articulatory imitation of facial movements during vocalization of a humanoid robot

In human heads there is a strong structural linkage between vocal tract and facial behavior during speech. For a robotic talking head to have a human-like behavior, this linkage should be emulated. One way to do that is to compute an estimate of the articulatory features which produce a given utterance and then to transform them into facial animation. We present a computational model of human vocalization which is aimed at describing the articulatory mechanisms which produce spoken phonemes. It uses a set of fuzzy rules and genetic optimization. The former represents the relationships between places of articulations and speech acoustic parameters, while the latter estimates the degrees of membership of the places of articulation. That is, the places of articulation are considered as fuzzy sets whose degrees of membership are the articulatory features. The trajectories of articulatory parameters can be used to control a graphical or mechanical talking head. We verify the model presented here by generating and listening to artificial sentences. Subjective listening tests of artificially generated sentences from the articulatory description resulted in an average phonetic accuracy of about 79 %. Through the analysis of a large amount of natural speech, the algorithm can be used to learn the places of articulation of all phonemes of a given speaker

Pro-active service robots in a health care framework: vocal interaction using natural language and prosody

This paper describes a service robotics application in a health care environment. Human-robot interaction is based on the integration of low level skills like acoustic localization of noise/voice sources with high level competencies like continuous speech recognition and prosody based dialogue. A multirobotic system is presented; its coordination is performed by a central supervision device which manages the vocal interaction with a number of human operators. A microphone array placed on the top of the robots is used both for near talker location and for auto-localization through acoustic beacons. Preliminary experimental results are discussed.

A Decision Support System for Cooperative Logistics

This paper specifies a cloud-based cooperative decision support system (DSS) that aims at integrating logistics management and decision support strategies for intermodal transportation systems. The proposed DSS is dedicated to synchronize different transportation means by using the modern information and communications technology tools and by taking into account environmental aspects. This paper describes the DSS cloud-based architecture and presents the procedure to be followed in order to design a DSS able to support decision makers in different logistic decision fields. The advantages of the proposed DSS are enlightened by specifying three main decision modules: cargo transport optimization, intelligent truck parking, and CO2 monitoring. Moreover, the applicability of the proposed DSS is described by specifying a DSS for the case study of the logistic network of the Trieste port (Italy), including the port, the inland terminal, and the highway connecting them. Some simulation campaigns are employed both to set the decision modules and evaluate the DSS application benefits.