Source author record

Carlos Gonzalez

Carlos Gonzalez appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Robotics astro-ph.IM astro-ph.SR Computer Vision gr-qc math-ph math.MP physics.class-ph physics.plasm-ph

Catalog footprint

What is connected

7works

9topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

Data-Driven Safety Verification for Legged Robots

Planning safe motions for legged robots requires sophisticated safety verification tools. However, designing such tools for such complex systems is challenging due to the nonlinear and high-dimensional nature of these systems' dynamics. In this letter, we present a probabilistic verification framework for legged systems, which evaluates the safety of planned trajectories by learning an assessment function from trajectories collected from a closed-loop system. Our approach does not require an analytic expression of the closed-loop dynamics, thus enabling safety verification of systems with complex models and controllers. Our framework consists of an offline stage that initializes a safety assessment function by simulating a nominal model and an online stage that adapts the function to address the sim-to-real gap. The performance of the proposed approach for safety verification is demonstrated using a quadruped balancing task and a humanoid reaching task. The results demonstrate that our framework accurately predicts the systems' safety both at the planning phase to generate robust trajectories and at execution phase to detect unexpected external disturbances.

preprint2022arXiv

Dictionary Learning with Accumulator Neurons

The Locally Competitive Algorithm (LCA) uses local competition between non-spiking leaky integrator neurons to infer sparse representations, allowing for potentially real-time execution on massively parallel neuromorphic architectures such as Intel's Loihi processor. Here, we focus on the problem of inferring sparse representations from streaming video using dictionaries of spatiotemporal features optimized in an unsupervised manner for sparse reconstruction. Non-spiking LCA has previously been used to achieve unsupervised learning of spatiotemporal dictionaries composed of convolutional kernels from raw, unlabeled video. We demonstrate how unsupervised dictionary learning with spiking LCA (\hbox{S-LCA}) can be efficiently implemented using accumulator neurons, which combine a conventional leaky-integrate-and-fire (\hbox{LIF}) spike generator with an additional state variable that is used to minimize the difference between the integrated input and the spiking output. We demonstrate dictionary learning across a wide range of dynamical regimes, from graded to intermittent spiking, for inferring sparse representations of both static images drawn from the CIFAR database as well as video frames captured from a DVS camera. On a classification task that requires identification of the suite from a deck of cards being rapidly flipped through as viewed by a DVS camera, we find essentially no degradation in performance as the LCA model used to infer sparse spatiotemporal representations migrates from graded to spiking. We conclude that accumulator neurons are likely to provide a powerful enabling component of future neuromorphic hardware for implementing online unsupervised learning of spatiotemporal dictionaries optimized for sparse reconstruction of streaming video from event based DVS cameras.

preprint2020arXiv

Line Walking and Balancing for Legged Robots with Point Feet

The ability of legged systems to traverse highly-constrained environments depends by and large on the performance of their motion and balance controllers. This paper presents a controller that excels in a scenario that most state-of-the-art balance controllers have not yet addressed: line walking, or walking on nearly null support regions. Our approach uses a low-dimensional virtual model (2-DoF) to generate balancing actions through a previously derived four-term balance controller and transforms them to the robot through a derived kinematic mapping. The capabilities of this controller are tested in simulation, where we show the 90kg quadruped robot HyQ crossing a bridge of only 6 cm width (compared to its 4 cm diameter spherical foot), by balancing on two feet at any time while moving along a line. Lastly, we present our preliminary experimental results showing HyQ balancing on two legs while being disturbed.

preprint2016arXiv

GIARPS: the unique VIS-NIR high precision radial velocity facility in this world

GIARPS (GIAno & haRPS) is a project devoted to have on the same focal station of the Telescopio Nazionale Galileo (TNG) both the high resolution spectrographs HARPS-N (VIS) and GIANO (NIR) working simultaneously. This could be considered the first and unique worldwide instrument providing cross-dispersed echelle spectroscopy at a high resolution (R=115,000 in the visual and R=50,000 in the IR) and over in a wide spectral range (0.383 - 2.45 micron) in a single exposure. The science case is very broad, given the versatility of such an instrument and the large wavelength range. A number of outstanding science cases encompassing mainly extra-solar planet science starting from rocky planet search and hot Jupiters, atmosphere characterization can be considered. Furthermore both instrument can measure high precision radial velocity by means the simultaneous thorium technique (HARPS - N) and absorbing cell technique (GIANO) in a single exposure. Other science cases are also possible. Young stars and proto-planetary disks, cool stars and stellar populations, moving minor bodies in the solar system, bursting young stellar objects, cataclysmic variables and X-ray binary transients in our Galaxy, supernovae up to gamma-ray bursts in the very distant and young Universe, can take advantage of the unicity of this facility both in terms of contemporaneous wide wavelength range and high resolution spectroscopy.

preprint2015arXiv

Power law inflation with a non-minimally coupled scalar field in light of Planck and BICEP2 data: The exact versus slow roll results

We study power law inflation in the context of non-minimally coupled to the scalar curvature. We analyze the inflationary solutions under an exact analysis and also in the slow roll approximation. In both solutions, we consider the recent data from Planck and BICEP2 data to constraint the parameter in our model. We find that the slow roll approximation is disfavored in the presence of non-minimal couplings during the power law expansion of the Universe.

preprint2014arXiv

Two-fluid turbulence including electron inertia

We present a full two-fluid magnetohydrodynamic (MHD) description for a completely ionized hydrogen plasma, retaining the effects of the Hall current, electron pressure and electron inertia. According to this description, each plasma species introduces a new spatial scale: the ion inertial length $λ_{i}$ and the electron inertial length $λ_{e}$, which are not present in the traditional MHD description. In the present paper, we seek for possible changes in the energy power spectrum in fully developed turbulent regimes, using numerical simulations of the two-fluid equations in two-and-a-half dimensions (2.5D). We have been able to reproduce different scaling laws in different spectral ranges, as it has been observed in the solar wind for the magnetic energy spectrum. At the smallest wavenumbers where plain MHD is valid, we obtain an inertial range following a Kolmogorov $k^{-5/3}$ law. For intermediate wavenumbers such that $λ_{i}^{-1} << k << λ_{e}^{-1}$, the spectrum is modified to a $k^{-7/3}$ power-law, as has also been obtained for Hall-MHD (HMHD) neglecting electron inertia terms. When electron inertia is retained, a new spectral region given by $k > λ_{e}^{-1}$ arises. The power spectrum for magnetic energy in this region is given by a $k^{-11/3}$ power law. Finally, when the terms of electron inertia are retained, we study the self-consistent electric field. Our results are discussed and compared with those obtained in solar wind observations and previous simulations.

preprint2011arXiv

Scattering solution of a ball by a bat

The problem of the mechanical evolution of a shock between a cylindrically symmetric bat and a spherical ball is solved in the strict rigid approximation for arbitrary values of the initial conditions. The friction during the impact is assumed to satisfy the standard rules. When the only source of energy dissipation is friction, the problem is fully solved by determining the separation point between the bodies. It also follows that whatever the character of any additional form of dissipation is, it only affects the ending value of the net impulse I done by the normal force of the bat on the ball at separation, but not the dynamical evolution with the value of I during the shock process. A relation determining whether the contact points of the two bodies slides between them or become at rest (to be pure rotation state) at the end of the impact, is determined for the case of the purely frictional energy dissipation. The solution is also generalized to include losses in addition to the frictional ones and then applied to the description of experimental measures of the scattering of a ball by a bat. The evaluations satisfactorily reproduce the measured curves for the output center of mass and angular velocities of the ball as functions of the scattering angle and the impact parameter, respectively.