First-person representations and responsible agency in AI.

In this paper I investigate which of the main conditions proposed in the moral responsibility literature are the ones that spell trouble for the idea that Artificial Intelligence Systems (AISs) could ever be full-fledged responsible agents. After arguing that the standard construals of the control and epistemic conditions don't impose any in-principle barrier to AISs being responsible agents, I identify the requirement that responsible agents must be aware of their own actions as the main locus of resistance to attribute that kind of agency to AISs. This is because this type of awareness is thought to involve first-person or de se representations, which, in turn, are usually assumed to involve some form of consciousness. I clarify what this widespread assumption involves and conclude that the possibility of AISs' moral responsibility hinges on what the correct theory of de se representations ultimately turns out to be.

Determination of Actual Friction Factors in Metal Forming under Heavy Loaded Regimes Combining Experimental and Numerical Analysis.

Tribological conditions can change drastically during heavy loaded regimes as experienced in metal forming; this is especially critical when lubrication can only be applied at the early stage of the process because the homogeneous lubricant layer can break along the die-workpiece interface. In these cases, adopting a constant friction factor for the lubricant-surface pair may not be a valid assumption. This paper presents a procedure based on the use of dual friction factor maps to determine friction factors employed in heavy loaded regimes. A finite element (FE) simulation is used to obtain the friction factor map for the alloy UNS A96082. Experiments were conducted using four lubricants (aluminum anti-size, MoS2 grease, silicone oil, and copper paste) to determine the actual friction curves. The experimental procedure is based on the application of lubricant only at the beginning of the first stage of ring compression, and not at intermediate stages as is usual in typical ring compression tests (RCTs). The results show that for small reductions (rh < 20%), the conventional RCT can be applied because the tribological conditions remain similar. For large reductions (rh > 20%), it is recommended to obtain an average value of the friction factor for every lubricant-surface pair in the range of deformation considered.

Surface roughness model based on force sensors for the prediction of the tool wear.

In this study, a methodology has been developed with the objective of evaluating the surface roughness obtained during turning processes by measuring the signals detected by a force sensor under the same cutting conditions. In this way, the surface quality achieved along the process is correlated to several parameters of the cutting forces (thrust forces, feed forces and cutting forces), so the effect that the tool wear causes on the surface roughness is evaluated. In a first step, the best cutting conditions (cutting parameters and radius of tool) for a certain quality surface requirement were found for pieces of UNS A97075. Next, with this selection a model of surface roughness based on the cutting forces was developed for different states of wear that simulate the behaviour of the tool throughout its life. The validation of this model reveals that it was effective for approximately 70% of the surface roughness values obtained.

An electrophysiological study of haptic roughness: effects of levels of texture and stimulus uncertainty in the P300.

This study investigated the neural mechanisms implicated in tactile perception using a discrimination task. We also investigated the influence of the type of presentation on the subject's uncertainty. The stimuli varied across four levels of roughness and were presented using a pure/mixed block design. We used an oddball paradigm with three target stimuli varying in the level of roughness, and a smooth surface as the non-target. Stimuli were presented using a specific-purpose device. We analyzed the modulation of the P300 amplitude elicited by targets and non-targets in both presentation conditions. The results showed that the P300 waveform was modulated by roughness, as well as by the order of stimuli presentation. The P300 amplitude was more sensitive to roughness when stimuli were presented in mixed blocks (higher uncertainty). The results are discussed in the context of the attention resources allocation theory applied to tactile modality.

Application of an instrumental and computational approach for improving the vibration behavior of structural panels using a lightweight multilayer composite.

This work presents a hybrid (experimental-computational) application for improving the vibration behavior of structural components using a lightweight multilayer composite. The vibration behavior of a flat steel plate has been improved by the gluing of a lightweight composite formed by a core of polyurethane foam and two paper mats placed on its faces. This composite enables the natural frequencies to be increased and the modal density of the plate to be reduced, moving about the natural frequencies of the plate out of excitation range, thereby improving the vibration behavior of the plate. A specific experimental model for measuring the Operating Deflection Shape (ODS) has been developed, which enables an evaluation of the goodness of the natural frequencies obtained with the computational model simulated by the finite element method (FEM). The model of composite + flat steel plate determined by FEM was used to conduct parametric study, and the most influential factors for 1st, 2nd and 3rd mode were identified using a multifactor analysis of variance (Multifactor-ANOVA). The presented results can be easily particularized for other cases, as it may be used in cycles of continuous improvement as well as in the product development at the material, piece, and complete-system levels.