Borghesan G.,Computer Science and Systems DEIS |
MacChelli A.,Computer Science and Systems DEIS |
Melchiorri C.,Computer Science and Systems DEIS
IEEE Transactions on Haptics | Year: 2010
In this paper, three results are presented concerning certain computational/control aspects, crucial for the proper behavior of haptic devices. The first one is a novel technique for a real-time simulation of virtual environments, which is able to preserve the energetic behavior of the simulated physical system and to avoid undesired effects related to unstable behaviors of the haptic device. The proposed real-time integration method is simpler, in terms of computational complexity, than similar solutions known in the literature, and provides an additional insight when faulty conditions are met. Second, a new method for the energy-consistent interconnection of discrete-time physical systems, implemented by algorithms running at different frequencies (i.e., multirate systems), is illustrated. Multirate systems are very common in haptics, since the frequency, at which the control law of the haptic interface is executed, is usually higher than the frequency of the simulation of the virtual environment. Finally, the third result presented in this paper concerns the problem of energy generation due to the time discretization in the acquisition of the haptic interface position. Similarly, to the previous case, a technique for an energy-consistent analog/digital conversion is proposed. All these methodologies have been validated, both by simulations and experiments. © 2010 IEEE.
Zavaglia M.,Computer Science and Systems DEIS |
Canolty R.T.,University of California at Berkeley |
Schofield T.M.,University College London |
Leff A.P.,University College London |
And 3 more authors.
Neural Networks | Year: 2012
This paper describes a dynamical process which serves both as a model of temporal pattern recognition in the brain and as a forward model of neuroimaging data. This process is considered at two separate levels of analysis: the algorithmic and implementation levels. At an algorithmic level, recognition is based on the use of Occurrence Time features. Using a speech digit database we show that for noisy recognition environments, these features rival standard cepstral coefficient features. At an implementation level, the model is defined using a Weakly Coupled Oscillator (WCO) framework and uses a transient synchronization mechanism to signal a recognition event. In a second set of experiments, we use the strength of the synchronization event to predict the high gamma (75-150Hz) activity produced by the brain in response to word versus non-word stimuli. Quantitative model fits allow us to make inferences about parameters governing pattern recognition dynamics in the brain. © 2012 Elsevier Ltd.