Search

This chapter reports an overview of the experience and the results achieved during the development of the robotized system for switchgear wiring carried out in the WIRES experiment. This specific application is particularly challenging for a robotic system due to the complexity of the manipulation task. As a matter of fact, in this task deformable linear objects, such as electric wires, are involved. Moreover, the precision requested during the assembly task and the typical crowded space inside the switchgear imply, on one side, the development of specific hardware and software tools and, on the other side, high adaptability and flexibility of the robotic system. In the WIRES experiment, a software package to extract the wiring information and to generate the robot task sequence directly from the switchgear CAD files has been developed. Additionally, a computer vision system able to recognize the location of the wires and of the electromechanical components inside the switchgear has been developed. To deal with the wire deformability and occlusion problems during the wire insertion, machine learning and sensor fusion techniques have been adopted to enable the wire manipulation by means of tactile sensors and 2D cameras feedback. From the hardware point of view, a specific end effector has been developed to manipulate and connect the wire to the components. This end effector is equipped with an electric screwdriver and a customized tactile sensor used to evaluate the wire shape, the wire end pose and its interaction with the environment during the manipulation. The entire task pipeline, going from the switchgear information extraction, to the wire grasp and manipulation, its connection and the routing along the desired wire path is presented in this chapter. The preliminary experimental results show that the developed system can achieve a success rate of about 95% in the wire insertion and connection task.

We propose a time-domain multiple sound source localization (SSL) method based on Bayesian inference. This method is specifically designed to run on the stochastic machines (SM) that we are currently developing to perform efficient low-level sensor signal processing with ultra-low power consumption. The proposed SSL method is divided into two main parts. First, a probabilistic model is run on 50 very short time frames (3. 75ms each) of multichannel recorded signals. Second, the results obtained on the different frames are fused to obtain a final localization map. Using the system in a supervised way allows to extract estimated source locations by selecting as many maxima as there are sources in the room. We explain how this method is implemented on a SM. Experiments are presented to illustrate the performance and robustness of the resulting system.

From an industrial and technological point of view, fuzzy control theory deals with the development of a particular system controller on a specific hardware by means of an open or legacy programming language that is useful to address, in a high-level fashion, the hardware constraints. Independently from the complexity of the addressed application problem, the development time may be very expensive if the designers want to exploit the possibility to collect different controllers in terms of an active, cooperative and heterogeneous virtual organization. In order to bridge this gap, we introduce the Fuzzy Markup Language (FML), a novel specific-purpose computer language that defines a detailed structure of fuzzy control independent from its legacy representation. This choice allows systems' designers to to express their ideas in fast and simple way and, as a consequence, speeds up the whole development process of a given complex system. © 2011 IEEE.

The centralized control of active distribution networks with large penetration of dispersed resources requires algorithms capable of providing network state estimation. Within this context, the measurement of bus voltage synchrophasors provides useful data that increase the observability of the network. The peculiar characteristics of distribution networks, such as feeder impedances, power flows and harmonic distortion, are different from those of transmission networks. Therefore, in order to obtain reasonably low uncertainty levels, Remote Terminal Units (RTU) with specific characteristics need to be developed. The paper presents an algorithm developed for the synchrophasor measurement in active distribution networks together with its implementation into a specific hardware based on a real-time microcontroller. The paper also presents the experimental characterization of the developed RTU by making reference to both single-tone and distorted signals. © 2009 IEEE.

This paper presents specific hardware and components which can be used to implement optimal control strategies via wireless communication channels to achieve path tracking. The position of the robot at each time instant Is found by implementing a localization algorithm. Simulations results are presented using robust and Linear Quadratic Gaussian (LQG) controller architectures. The overall aim of this research is the integration of the controller in a communication system where the sensor and/or reference information is transmitted via a channel with finite capacity. © 2008 IEEE. 946 951

Modern surveillance systems, such as those based on the use of Unmanned Aerial Vehicles, require powerful high- performance platforms to deal with many different algorithms that make use of massive calculations. At the same time, low- cost and high-performance specific hardware (e.g., GPU, PPU) are rising and the CPUs turned to multiple cores, characteriz- ing together an interesting and powerful heterogeneous execu- tion platform. Therefore, reconfigurable computing is a poten- tial paradigm for those scenarios as it can provide flexibility to explore the computational resources on heterogeneous cluster attached to a high-performance computer system platform. As the first step towards a run-time reconfigurable workload bal- ancing framework targeting that kind of platform, application time requirements and its crosscutting behavior play an impor- tant role for task allocation decisions. This paper presents a strategy to reallocate specific tasks in a surveillance system composed by a fleet of Unmanned Aerial Vehicles using aspect- oriented paradigms in order to address non-functional applica- tion timing constraints in the design phase. An aspect support from a framework called DERAF is used to support reconfigu- ration requirements and provide the resource information needed by the reconfigurable load-balancing strategy. Finally, for the case study, a special attention on Radar Image Process- ing will be given.