Heuristic approaches for the optimal wiring in large scale robotic skin design

This paper faces the problem of optimizing the wiring and the connections in a tactile skin for robots. The robotic skin is a device composed of a network of tactile sensors, whose wiring can be very complex: the control of this complexity is a key problem. In the considered robotic skin, skin elements are grouped into skin patches, which output tactile data that have to be read by a micro-controller. The logical connections between the sensors must be defined in order to route signals through the network. A finite set of micro-controllers is given and a set of constraints is imposed on the given assignment and routing. The considered problem has a combinatorial nature and it can be formulated as a Minimum Constrained Spanning Forest problem with costs on arcs that cannot be a priori defined as they are solution-dependent. The problem is NP-hard. The paper introduces a mathematical formulation and then proposes a Multi-Start Heuristic algorithm and an Ant Colony Optimization approach whose effectiveness is evaluated through experimental tests performed on both real and synthetically generated instances.     

Feedforward networks in financial predictions: the future that modifies the present

The main goal of this paper is to show how relatively minor modifications of well-known algorithms (in particular, back propagation) can dramatically increase the performance of an artificial neural network (ANN) for time series prediction. We denote our proposed sets of modifications as the 'self-momentum', 'Freud' and 'Jung' rules. In our opinion, they provide an example of an alternative approach to the design of learning strategies for ANNs, one that focuses on basic mathematical conceptualization rather than on formalism and demonstration. The complexity of actual prediction problems makes it necessary to experiment with modelling possibilities whose inherent mathematical properties are often not well understood yet. The problem of time series prediction in stock markets is a case in point. It is well known that asset price dynamics in financial markets are difficult to trace, let alone to predict with an operationally interesting degree of accuracy. We therefore take financial prediction as a meaningful test bed for the validation of our techniques. We discuss in some detail both the theoretical underpinnings of the technique and our case study about financial prediction, finding encouraging evidence that supports the theoretical and operational viability of our new ANN specifications. Ours is clearly only a preliminary step. Further developments of ANN architectures with more and more sophisticated 'learning to learn' characteristics are now under study and test.     

Evaluating an augmented reality interface for sailing navigation: a comparative study with a immersive virtual reality simulator

Sailing navigation is an activity that requires acquiring and processing information from the surrounding environment. The advancement of technology has enabled sailboats to have an increasing number of onboard sensors that make sailing more user-friendly. However, data provided by these sensors are still visualized on 2D digital displays that imitate traditional analog interfaces. Although these displays are strategically placed on the sailboat, the user needs to divert attention from the primary navigation task to look at them, thus spending a significant amount of cognitive resources. AR-based technologies have the potential to overcome these limitations by displaying information registered in the real environment, but there are no studies in the literature for validating the effectiveness of this technology in the field of sailing. Thus, we designed a head-mounted display AR-based interface to assist users in monitoring wind data to avoid user diversion from the primary task of sailing. We conducted a user study involving 45 participants in an Immersive Virtual Reality simulated environment. We collected objective and subjective measures to compare the AR-based interface with a traditional data visualization system. The AR-based interface outperformed the traditional data visualization system regarding reaction time, cognitive load, system usability, and user experience.

     

A hybrid positioning system for technology‐independent location‐aware computing

Location‐aware computing is a form of context‐aware mobile computing that refers to the ability of providing users with services that depend on their position. Locating the user terminal, often called positioning, is essential in this form of computing. Towards this aim, several technologies exist, ranging from personal area networking, to indoor, outdoor, and up to geographic area systems. Developers of location‐aware software applications have to face with a number of design choices, that typically depend on the chosen technology. This work addresses the problem of easing the development of pull location‐aware applications, by allowing uniform access to multiple heterogeneous positioning systems. Towards this aim, the paper proposes an approach to structure location‐aware mobile computing systems in a way independent of positioning technologies. The approach consists in structuring the system into a layered architecture, that provides application developers with a standard Java Application Programming Interface (JSR‐179 API), and encapsulates location data management and technology‐specific positioning subsystems into lower layers with clear interfaces. In order to demonstrate the proposed approach we present the development of HyLocSys. It is an open hybrid software architecture designed to support indoor/outdoor applications, which allows the uniform (combined or separate) use of several positioning technologies. HyLocSys uses a hybrid data model, which allows the integration of different location information representations (using symbolic and geometric coordinates). Moreover, it allows support to handset‐ and infrastructure‐based positioning approaches while respecting the privacy of the user. The paper presents a prototypal implementation of HyLocSys for heterogeneous scenarios. It has been implemented and tested on several platforms and mobile devices. Copyright © 2009 John Wiley & Sons, Ltd.     

Efficient breadth first search on multi-GPU systems

Simple algorithms for the execution of a Breadth First Search on large graphs lead, running on clusters of GPUs, to a situation of load unbalance among threads and un-coalesced memory accesses, resulting in pretty low performances. To obtain a significant improvement on a single GPU and to scale by using multiple GPUs, we resort to a suitable combination of operations to rearrange data before processing them. We propose a novel technique for mapping threads to data that achieves a perfect load balance by leveraging prefix-sum and binary search operations. To reduce the communication overhead, we perform a pruning operation on the set of edges that needs to be exchanged at each BFS level. The result is an algorithm that exploits at its best the parallelism available on a single GPU and minimizes communication among GPUs. We show that a cluster of GPUs can efficiently perform a distributed BFS on graphs with billions of nodes.     

Using recurrent neural networks to detect changes in autocorrelated processes for quality monitoring

With the growing of automation in manufacturing, process quality characteristics are being measured at higher rates and data are more likely to be autocorrelated. A widely used approach for statistical process monitoring in the case of autocorrelated data is the residual chart. This chart requires that a suitable model has been identified for the time series of process observations before residuals can be obtained. In this work, a new neural-based procedure, which is alleviated from the need for building a time series model, is introduced for quality control in the case of serially correlated data. In particular, the Elman’s recurrent neural network is proposed for manufacturing process quality control. Performance comparisons between the neural-based algorithm and several control charts are also presented in the paper in order to validate the approach. Different magnitudes of the process mean shift, under the presence of various levels of autocorrelation, are considered. The simulation results indicate that the neural-based procedure may perform better than other control charting schemes in several instances for both small and large shifts. Given the simplicity of the proposed neural network and its adaptability, this approach is proved from simulation experiments to be a feasible alternative for quality monitoring in the case of autocorrelated process data.     

Knowledge Management in Locating the Patient in an Emergency Medical Service in Italy

This study examines an Emergency Medical Service in order to analyze the composite set of activities and instruments directed at locating the patient. The good management of information about the location of the emergency is highly relevant for a reliable rescue service, but this information depends on knowledge of the territory that is socially distributed between EMS operators and callers. Accordingly, the decision-making process often has to go beyond the emergency service protocols, engaging the operator in undertaking an open negotiation in order to transform the caller’s role from layman to “co-worker”. The patient’s location turns out to be an emerging phenomenon, collaborative work based on knowledge management involving two communities—the callers and the EMS operators—that overlap partially. Drawing examples from emergency calls, the study analyzes the practice of locating a patient as a complex and multi-layered process, highlighting the role played by new and old technologies (the information system and the paper maps) in this activity. We argue that CSCW technologies enable the blended use of different kinds of instruments and support an original interconnection between the professional localization systems and the public’s way of defining a position.     

Visual routines for real-time monitoring of vehicle behavior

Low-cost, purely visual modules are presented, which are able to reliably perform real-time computation of kinematic quantities relevant for the navigation of commercial vehicles moving along highways or motorways. Taking as input a video stream of b/w images coming from a standard camera mounted aboard the vehicle, the modules return instantaneous measurements for the speed of the vehicle, its lateral position in the lane, and the distance from the preceding vehicle. The modules, which are designed to work in real-time on standard PC platforms, have been tested under typical working conditions. Preliminary results are reported and discussed. Received: 10 March 1997 / Accepted: 24 November 1997