Search

Semiconductor nanowires can be utilized to create quantum dot qubits. The formation of quantum dots is typically achieved by means of bottom gates created by a lift-off process. As an alternative, we fabricated flat buried bottom gate structures by filling etched trenches in a Si substrate with sputtered TiN, followed by mechanical polishing. This method achieved gate line pitches as small as 60 nm. The gate fingers have low gate leakage. As a proof of principle, we fabricated quantum dot devices using InAs nanowires placed on the gate fingers. These devices exhibit single electron tunneling and Coulomb blockade., Comment: 7 pages main manuscript, 5 pages supplementary material, 6 figures in main manuscript, 6 supplementary figures

We investigate the operation of dual-gate reconfigurable field-effect transistor (RFET) in the programgate at drain (PGAD) and program-gate at source (PGAS) configurations. To this end, dual-gate silicon nanowire (SiNW) FETs are fabricated based on anisotropic wet chemical silicon etching and nickel silicidation yielding silicide-SiNW Schottky junctions at source and drain. Whereas in PGAD-configuration ambipolar operation is suppressed, switching is deteriorated due to the injection through a Schottky-barrier. Operating the RFET in PGAS configuration yields a switching behavior close to a conventional MOSFET. This, howewer, needs to be traded off against strongly non-linear output characteristics for small bias., Comment: 5 pages

The aim of this paper is to map agricultural crops by classifying satellite image time series. Domain experts in agriculture work with crop type labels that are organised in a hierarchical tree structure, where coarse classes (like orchards) are subdivided into finer ones (like apples, pears, vines, etc.). We develop a crop classification method that exploits this expert knowledge and significantly improves the mapping of rare crop types. The three-level label hierarchy is encoded in a convolutional, recurrent neural network (convRNN), such that for each pixel the model predicts three labels at different level of granularity. This end-to-end trainable, hierarchical network architecture allows the model to learn joint feature representations of rare classes (e.g., apples, pears) at a coarser level (e.g., orchard), thereby boosting classification performance at the fine-grained level. Additionally, labelling at different granularity also makes it possible to adjust the output according to the classification scores; as coarser labels with high confidence are sometimes more useful for agricultural practice than fine-grained but very uncertain labels. We validate the proposed method on a new, large dataset that we make public. ZueriCrop covers an area of 50 km x 48 km in the Swiss cantons of Zurich and Thurgau with a total of 116'000 individual fields spanning 48 crop classes, and 28,000 (multi-temporal) image patches from Sentinel-2. We compare our proposed hierarchical convRNN model with several baselines, including methods designed for imbalanced class distributions. The hierarchical approach performs superior by at least 9.9 percentage points in F1-score.

The application of autonomous robots in agriculture is gaining increasing popularity thanks to the high impact it may have on food security, sustainability, resource use efficiency, reduction of chemical treatments, and the optimization of human effort and yield. With this vision, the Flourish research project aimed to develop an adaptable robotic solution for precision farming that combines the aerial survey capabilities of small autonomous unmanned aerial vehicles (UAVs) with targeted intervention performed by multi-purpose unmanned ground vehicles (UGVs). This paper presents an overview of the scientific and technological advances and outcomes obtained in the project. We introduce multi-spectral perception algorithms and aerial and ground-based systems developed for monitoring crop density, weed pressure, crop nitrogen nutrition status, and to accurately classify and locate weeds. We then introduce the navigation and mapping systems tailored to our robots in the agricultural environment, as well as the modules for collaborative mapping. We finally present the ground intervention hardware, software solutions, and interfaces we implemented and tested in different field conditions and with different crops. We describe a real use case in which a UAV collaborates with a UGV to monitor the field and to perform selective spraying without human intervention., Comment: Published in IEEE Robotics & Automation Magazine, vol. 28, no. 3, pp. 29-49, Sept. 2021

We present a novel weed segmentation and mapping framework that processes multispectral images obtained from an unmanned aerial vehicle (UAV) using a deep neural network (DNN). Most studies on crop/weed semantic segmentation only consider single images for processing and classification. Images taken by UAVs often cover only a few hundred square meters with either color only or color and near-infrared (NIR) channels. Computing a single large and accurate vegetation map (e.g., crop/weed) using a DNN is non-trivial due to difficulties arising from: (1) limited ground sample distances (GSDs) in high-altitude datasets, (2) sacrificed resolution resulting from downsampling high-fidelity images, and (3) multispectral image alignment. To address these issues, we adopt a stand sliding window approach that operates on only small portions of multispectral orthomosaic maps (tiles), which are channel-wise aligned and calibrated radiometrically across the entire map. We define the tile size to be the same as that of the DNN input to avoid resolution loss. Compared to our baseline model (i.e., SegNet with 3 channel RGB inputs) yielding an area under the curve (AUC) of [background=0.607, crop=0.681, weed=0.576], our proposed model with 9 input channels achieves [0.839, 0.863, 0.782]. Additionally, we provide an extensive analysis of 20 trained models, both qualitatively and quantitatively, in order to evaluate the effects of varying input channels and tunable network hyperparameters. Furthermore, we release a large sugar beet/weed aerial dataset with expertly guided annotations for further research in the fields of remote sensing, precision agriculture, and agricultural robotics., Comment: 25 pages, 14 figures, MDPI Remote Sensing

Selective weed treatment is a critical step in autonomous crop management as related to crop health and yield. However, a key challenge is reliable, and accurate weed detection to minimize damage to surrounding plants. In this paper, we present an approach for dense semantic weed classification with multispectral images collected by a micro aerial vehicle (MAV). We use the recently developed encoder-decoder cascaded Convolutional Neural Network (CNN), Segnet, that infers dense semantic classes while allowing any number of input image channels and class balancing with our sugar beet and weed datasets. To obtain training datasets, we established an experimental field with varying herbicide levels resulting in field plots containing only either crop or weed, enabling us to use the Normalized Difference Vegetation Index (NDVI) as a distinguishable feature for automatic ground truth generation. We train 6 models with different numbers of input channels and condition (fine-tune) it to achieve about 0.8 F1-score and 0.78 Area Under the Curve (AUC) classification metrics. For model deployment, an embedded GPU system (Jetson TX2) is tested for MAV integration. Dataset used in this paper is released to support the community and future work.

We discuss the density shift and broadening of Rydberg spectra measured in cold, dense atom clouds in the context of Rydberg atom spectroscopy done at room temperature, dating back to the experiments of Amaldi and Segr\`e in 1934. We discuss the theory first developed in 1934 by Fermi to model the mean-field density shift and subsequent developments of the theoretical understanding since then. In particular, we present a model whereby the density shift is calculated using a microscopic model in which the configurations of the perturber atoms within the Rydberg orbit are considered. We present spectroscopic measurements of a Rydberg atom, taken in a Bose-Einstein condensate (BEC) and thermal clouds with densities varying from $5\times10^{14}\textrm{cm}^{-3}$ to $9\times10^{12}\textrm{cm}^{-3}$. The density shift measured via the spectrum's center of gravity is compared with the mean-field energy shift expected for the effective atom cloud density determined via a time of flight image. Lastly, we present calculations and data demonstrating the ability of localizing the Rydberg excitation via the density shift within a particular density shell for high principal quantum numbers., Comment: 27 pages, 12 figures, Review Paper

Within a dense environment ($\rho \approx 10^{14}\,$atoms/cm$^3$) at ultracold temperatures ($T < 1\,\mu{}\text{K}$), a single atom excited to a Rydberg state acts as a reaction center for surrounding neutral atoms. At these temperatures almost all neutral atoms within the Rydberg orbit are bound to the Rydberg core and interact with the Rydberg atom. We have studied the reaction rate and products for $nS$ $^{87}$Rb Rydberg states and we mainly observe a state change of the Rydberg electron to a high orbital angular momentum $l$, with the released energy being converted into kinetic energy of the Rydberg atom. Unexpectedly, the measurements show a threshold behavior at $n\approx 100$ for the inelastic collision time leading to increased lifetimes of the Rydberg state independent of the densities investigated. Even at very high densities ($\rho\approx4.8\times 10^{14}\,\text{cm}^{-3}$), the lifetime of a Rydberg atom exceeds $10\,\mu\text{s}$ at $n > 140$ compared to $1\,\mu\text{s}$ at $n=90$. In addition, a second observed reaction mechanism, namely Rb$_2^+$ molecule formation, was studied. Both reaction products are equally probable for $n=40$ but the fraction of Rb$_2^+$ created drops to below 10$\,$% for $n\ge90$., Comment: 13 pages, 13 figures

We present spectroscopy of a single Rydberg atom excited within a Bose-Einstein condensate. We not only observe the density shift as discovered by Amaldi and Segre in 1934, but a line shape which changes with the principal quantum number n. The line broadening depends precisely on the interaction potential energy curves of the Rydberg electron with the neutral atom perturbers. In particular, we show the relevance of the triplet p-wave shape resonance in the Rydberg electron-Rb(5S) scattering, which significantly modifies the interaction potential. With a peak density of 5.5x10^14 cm^-3, and therefore an inter-particle spacing of 1300 a0 within a Bose-Einstein condensate, the potential energy curves can be probed at these Rydberg ion - neutral atom separations. We present a simple microscopic model for the spectroscopic line shape by treating the atoms overlapped with the Rydberg orbit as zero-velocity, uncorrelated, point-like particles, with binding energies associated with their ion-neutral separation, and good agreement is found., Comment: 5 pages, 4 figures

We present high-resolution spectroscopy of Rb$_\text{2}$ ultralong-range Rydberg molecules bound by mixed singlet-triplet electron-neutral atom scattering. The mixing of the scattering channels is a consequence of the hyperfine interaction in the ground-state atom, as predicted recently by Anderson et al. \cite{Anderson2014b}. Our experimental data enables the determination of the effective zero-energy singlet $s$-wave scattering length for Rb. We show that an external magnetic field can tune the contributions of the singlet and the triplet scattering channels and therefore the binding energies of the observed molecules. This mixing of molecular states via the magnetic field results in observed shifts of the molecular line which differ from the Zeeman shift of the asymptotic atomic states. Finally, we calculate molecular potentials using a full diagonalization approach including the $p$-wave contribution and all orders in the relative momentum $k$, and compare the obtained molecular binding energies to the experimental data.

We present a novel spectroscopic method for probing the \insitu~density of quantum gases. We exploit the density-dependent energy shift of highly excited {Rydberg} states, which is of the order $10$\MHz\,/\,1E14\,cm$^{\text{-3}}$ for \rubidium~for triplet s-wave scattering. The energy shift combined with a density gradient can be used to localize Rydberg atoms in density shells with a spatial resolution less than optical wavelengths, as demonstrated by scanning the excitation laser spatially across the density distribution. We use this Rydberg spectroscopy to measure the mean density addressed by the Rydberg excitation lasers, and to monitor the phase transition from a thermal gas to a Bose-Einstein condensate (BEC)., Comment: This paper omitted consideration of the p-wave shape resonance leading to an inaccurate understanding of the spectroscopic line broadening and ultimately the resolution with which low-n Rydberg atoms could be spatially localized. A thorough theoretical analysis of this data set is found now as arXiv:1607.01325, (to appear in JPhysB) and M. Schlagm\"uller et. al PRL 116 053001 (2016)

We present the first experimental study that maps the transformation of nuclear quadrupole resonances from the pure nuclear quadrupole regime to the quadrupole-perturbed Zeeman regime. The transformation presents an interesting quantum-mechanical problem, since the quantization axis changes from being aligned along the axis of the electric-field gradient tensor to being aligned along the magnetic field. We achieve large nuclear quadrupole shifts for I = 3/2 131-Xe by using a 1 mm^3 cubic cell with walls of different materials. When the magnetic and quadrupolar interactions are of comparable size, perturbation theory is not suitable for calculating the transition energies. Rather than use perturbation theory, we compare our data to theoretical calculations using a Liouvillian approach and find excellent agreement., Comment: 4 pages, 4 figures