Integrated Computer-Aided Engineering

Signal processing for slip and contact sensing for a robotic tactile sensor is presented. Small strips of piezoelectric film (PVDF) were embedded in the fingertip of a robotic finger made of silicon rubber. The sensor detected the vibration caused by slip and contact. A short impulsive vibration was generated when the finger made contact with an object, and a random noise-like signal was observed when slip occurred at the fingertip. Slip was differentiated from contact by counting the number of peaks in the waveform within a 6.4∼ms interval. This non-linear filtering method was shown to be more effective than ordinary frequency domain filtering. Several experiments using the two-fingered robotic hand are presented to demonstrate the importance of slip sensing in manipulative motion.

The Robotic systems for disabled people, aims at bringing a piloting assistance to powered wheelchairs in answer to the needs of mobility aid using dynamic vision techniques from mobile robotics. People start to design the robot with the fastest computers to process the information, together with high quality camera to serve as its eye, and precise mechanical motion control. The main working mode for this purpose is a contribution to dynamic vision, in the aim to improve a visio-space behavior of handicapped children. A self-calibration technique of a visual sensor is described: self-calibration implying the adoption of methods allowing to calibrate automatically a camera without using of special calibration set-ups. This paper examines what can be done within a Euclidean calibration, when the internal parameters must remain constant any more. So, we look at a Euclidean basis being a projective and one where some constraints must be observed. This work shows theoretically along with real experiments, how it is possible to completely calibrate a camera in line, that is to determine the intrinsic parameters and the relative displacement between two or three images, without any a priori knowledge of the scenes. The infinite homography computing method is used to estimate intrinsic and extrinsic parameters. This procedure consists of a closed-form solution, followed by a nonlinear refinement based on a bundle adjustment criterion.

This paper proposes an original integrated approach allowing coordination rules to be described for resource allocation in Distributed Multimedia Applications. Our approach has two major benefits. On the one hand, it allows the scalability and the dynamicity required for today's distributed systems. On the other hand, rather than bringing a new formalism and rather than making a dialect of an existing formal language we work on how two existing description techniques may be integrated to compositionally describe the whole system. An implementation of our approach is presented.

The paper presents the use of inductive machine learning for selecting appropriate features capable of detecting washing machines that have mechanical defects or that are wrongly assembled in the production line. The input data are vibration signals from different points of washing machine's surface when it operates for a couple of minutes in centrifuge mode. The signal is normalized and transformed to the frequency domain using Fast Fourier Transforms (FFT). An adequate example database is constructed from samples of different machine status. Our approach lends basic concepts contained in the algorithmic family of ID3 and C4.5. Certain parts of the algorithmic process contained in the above machine learning tools are used in order to construct a feature selection methodology based on information entropy criteria. The selected features are then used by specific classification techniques for achieving successful discrimination among different types of fault and normal operation. The overall methodology, using real data acquired within the European Community funded project called MEDEA, obtains a high rate of fault detection exceeding 87% of successful classification.

This paper addresses vision processing characteristics through a survey of parallel dedicated architectures and proposes a concept of vision application-adaptable architecture and its implementation called FreeTIV computer (Free architecture dedicated to image processing and vision).

FreeTIV is a heterogeneous parallel computer defined as a set of functionally reconfigurable processing modules. A message passing router called RouTIV (Router dedicated to image processing and vision) adapts FreeTIV module interconnections according to the running application data and the global application execution costs.

The adaptable computer concept allows to develop fast and reliable (dedicated or not) parallel computer at low cost.

In this paper, we propose a new complex-type wavelet which we call RI-Spline wavelet. We show that the RI-Spline wavelet satisfies the admissibility condition and bi-orthogonal condition. That is, the RI-Spline wavelet can be used for continuous and discrete wavelet transform as the mother wavelet. As an application of the RI-Spline wavelet, we examine the relationship between the sound and vibration signals produced by golf club head impact, and analyze characteristic of unsteady turbulence. Our results demonstrate the advantages of our approach and show that the RI-Spline wavelet is effective for nonstationary signal analysis.

Biotechnical and food process industries have a need for tools that support the implementation of modern manufacturing practice. Flexible production strategies and total quality management require efficient design of the process recipes.

This paper presents the approach used in Sophist, an adaptive recipe planning system. The system integrates case-based reasoning and a semi-qualitative process model for the design and optimisation of bioprocess recipes. In particular, we introduce a novel revision selection algorithm that improve the system's adaptivity. The behavior of the planner is exemplified by a batch fermentation planning experiment.