Abstract|This paper analyzes properties of a certain class of approximation techniques { HyperBF networks { in face perception tasks. The problem of gender classi cation and identi cation is addressed using a geometrical description of faces, extracted automatically from digitized pictures of frontal views of people without facial hair. The HyperBF networks perform satisfactorily on the classi cation tasks and exhibit the phenomenon of caricaturing, previously reported in psychophysical experiments. 1.

In the framework of autonomous navigation, an application of Kalman filtering to the problem of multi-sensor information processing is presented. In particular, estimation of model parameters is considered when a mobile robot equipped with a set of sonars and a standard TV camera moves along corridors, and the environment is affected by sharp discontinuities due to the presence of recesses or protrusions. Experiments performed in realworld situations are also presented and discussed. 1 INTRODUCTION Endowing a mobile vehicle with the capability of moving purposefully, autonomously and reliably in indoor environments is a field of research of evident technological interest, and a demanding test-bed for a variety of pure and applied disciplines. Performances an autonomous navigation system is able to achieve depend on several intertwined factors -- from computing power to sensors, to hardware and software tools, to planning and reasoning resources. In the last few years actual relevance ...

This paper deals with the architectural aspects of mobile robots interacting with external agents as well as with real environments. The problem is how to organize the large and flexible "gray area" left in between the reasoning capabilities and the sensorimotors activities. Our approach is based on objects we call "Basic Behaviours", fairly complex actions, constructed bottom-up from the sensorimotor level, as representationless reactions and activities; with appropriate descriptions, Basic Behaviours represent, for the symbolic level, units on which to reason. An architecture structured in three layers is then introduced. A number of relevant remarks and issues that arose during development and experimentation of our actual system is finally collected. 1 Introduction Research on mobile robots concerns the realization of vehicles able to move autonomously and reliably through unstructured, real-world environments, performing a useful mission and interacting with human operato...

Simple indoor navigation subtasks, such as moving an autonomous platform robot in a corridor parallel to a wall or correcting its trajectory to avoid small obstacles, can be accomplished using reflexive behaviours, without the need of a navigation planner. We describe a dynamic vision module in which reflexes are implemented as feedback systems. Measurements on the image data provide the position of the floor boundary in the coordinate system of the onboard cameras. The system state vector, estimated from the image measurements, represents the position of the platform relative to the walls and is used to compute the corrective commands for the actuators. The system, 2 implemented using standard hardware, has been tested in several conditions including driving in small areas or at high speeds. Introduction The I.R.S.T. autonomous navigation project aims to explore an architecture for indoor navigation and control that consists of several sensor-motor "reflexes ", implemented as class...

This paper describes a tool for the development of distributed applications. The target execution environment is a network of heterogenous general-purpose workstations. This tool (called HERMES) mainly handles the message transport layer and does not deal with aspects more directly related to parallel computing, such as the topology of communicating processes or load balancing among processors. A coarse-grained concurrence is supported and the resulting parallel architecture is based on the multiple-instructions multiple-data paradigm. This system is mainly characterized by its ability of handling special messages (alarm and urgent) that are guaranteed immediate processing by the receiving process. Messages are always asynchronous, in that they are immediately delivered to destination, without any temporary buffering by the communication system (i.e. asynchronously with respect to the program execution). The system is still under development but a preliminary version, restricted to pro...

The main purpose of MAIA, acronym for "Modello Avanzato di Intelligenza Artificiale" (Advanced Model of Artificial Intelligence), is to provide a common experimental platform for several research fields of Artificial Intelligence, such as Computer Vision, Automatic Speech Recognition, Natural Language Processing, Mechanized Reasoning, Human Factors, Knowledge Representation, Hardware and Software Architecture. A visitor of IRST is accompanied by a mobile robot to the desired destination and during the navigation is informed about research activities carried out in some places visited by the robot during transfer. If unpredictable events occur during this mission, caused by human presence or dynamic environment variation, the robot asks questions for continuation or replanning of mission. A prototype of the demonstration system has been fully implemented on a network of workstations. The first results are encouraging and the system is now under test. 1 Introduction MAIA ( Advanced Mode...

Research work reported in the present paper concerns the development of a navigation system for a mobile robot designed to accomplish missions inside the building of big organizations, such as hospitals or malls. Useful terminology is first introduced, and a rather detailed presentation of architecture and overall functioning of the system follows thereafter. The most recent realization of the system is then described in some detail, and behaviours and functionalities it can exhibit are reported and discussed. Possible directions for future work are finally outlined, along with open problems that are regarded as the most interesting and challenging for the further development of the system. I Introduction Realization of artificial systems able to move autonomously, purposefully and reliably in fairly unstructured indoor environments is an area of research of relevant technological impact, and may represent a severe testbed for a variety of pure and applied disciplines [22]. It is no...