A5101460738- Advanced Control Systems Optimization
- Control Systems and Identification
- Fault Detection and Control Systems
- Target Tracking and Data Fusion in Sensor Networks
- Parallel Computing and Optimization Techniques
- Radiation Effects in Electronics
- Probabilistic and Robust Engineering Design
- Elasticity and Wave Propagation
- Advanced Statistical Process Monitoring
- Software System Performance and Reliability
- Advanced Neural Network Applications
- Cognitive Science and Mapping
- Adversarial Robustness in Machine Learning
- Computational Physics and Python Applications
- Hydrological Forecasting Using AI
- Music Technology and Sound Studies
- Stochastic Gradient Optimization Techniques
- Gaussian Processes and Bayesian Inference
- Speech and Audio Processing
- Embedded Systems Design Techniques
- Leadership, Behavior, and Decision-Making Studies
- Multi-Criteria Decision Making
- Interconnection Networks and Systems
- Competitive and Knowledge Intelligence
- Structural mechanics and materials
Mondragon University
2024
Universitat Politècnica de Catalunya
2011-2022
GAIKER Technology Centre
2021-2022
Ikerlan
2021-2022
University of California, Los Angeles
2010-2015
Universidad Complutense de Madrid
2000
An efficient recursive state estimator is developed for two-state linear systems driven by Cauchy distributed process and measurement noises. For a general vector-state system, the based on recursively propagating characteristic function of conditional probability density (cpdf), where number terms in sum that expresses this grows with each update. Both mean error variance are functions history. two states, proposed reduces substantially needed to express cpdf taking advantage relationships...
An optimal predictive controller for linear, vector-state dynamic systems driven by Cauchy measurement and process noises is developed. For the system, probability distribution function (pdf) of state conditioned on history cannot be generated. However, characteristic this pdf can expressed in an analytic form. Consequently, performance index evaluated spectral domain using function. By objective that a product functions resembling pdfs, conditional obtained analytically closed form...
An optimal predictive controller for linear, vector-state dynamic systems driven by Cauchy measurement and process noises is developed. For the system, only characteristic function of conditional probability density (pdf), not pdf itself, can be expressed analytically in a closed form. Consequently, performance index formulated design also evaluated spectral domain. In particular, taking expectation an objective that product functions resembling pdfs, obtained form using Parseval's identity...
A multi-step predictive optimal control scheme is developed for scalar discrete linear dynamic systems driven by Cauchy distributed process and measurement noises. Although the densities that model noise have an undefined first moment infinite second moment, probability density function conditioned on noisy measurements does a finite mean variance. For problem cost criterion should be defined which unconditional expectation of this with respect to exists. The chosen functionally similar...
The conditional mean estimator for a two-state linear system with additive Cauchy measurement and process noises is developed. Although the densities that model noise have an undefined first moment infinite second moment, probability density function conditioned on noisy measurements does finite variance. (cpdf) given history appears to be difficult compute directly. However, characteristic of unnormalized cpdf can sequentially propagated through updates dynamic state propagation. A key step...
Deep learning technology has enabled the development of increasingly complex safety-related autonomous systems using high-performance computers, such as graphics processing units (GPUs), which provide required high computing performance for execution parallel algorithms, matrix–matrix multiplications (a central element deep software libraries). However, safety certification algorithms and GPU-based is a challenge to be addressed. For example, achieving fault-tolerance diagnostic coverage...
With the advent of next-generation safety-related systems, different industries face multiple challenges in ensuring safe operation these systems according to traditional safety and assurance techniques. The increasing complexity that characterizes hampers maximum achievable test coverage during system verification and, consequently, it often results untested behaviors hinder represent potential risk sources operation. In context paving way towards quantifying risks caused by software...
The ability of CNNs to efficiently and accurately perform complex functions, such as object detection, has fostered their adoption in safety-related autonomous systems. These algorithms require high computational performance platforms that exploit levels parallelism. control mitigation random errors these underlying become a must according functional safety standards. In this paper, we propose protecting, with catalog diagnostic techniques, the most computationally expensive operation CNNs,...