Ibeo Automotive Systems (Germany)

Deploying modern applications with significant resource demands is often challenging, but approximate designs offer a promising alternative by delivering high performance with minimal compromises in output quality. Traditionally, approximate hardware accelerators have been developed through search-based iterative frameworks, which suffer from long runtimes due to the exponential growth of the design space. A significant portion of the runtime is consumed by either invalid nodes or valid nodes that offer minimal improvements in performance metrics, such as runtime or power consumption. This severely limits the thorough exploration of the design space. In this paper, we introduce a novel approach for synthesizing approximate accelerators that leverages sparsity to reduce the complexity of the design space exploration problem. Our method employs a generative adversarial network (GAN) to rapidly generate a diverse set of high-quality design nodes, eliminating the need for costly node evaluations. This enables the swift creation of approximate accelerators generated for any given error threshold in a fraction of time as compared to a simulation-based framework. We conducted experiments on a suite of benchmarks from real-world domains, demonstrating that our methodology can generate approximate hardware designs with significant area and power savings, comparable to state-of-the-art search-based approaches. In a comparative evaluation against two leading methods, our approach achieved equal or better quality results for two out of four benchmarks while reaching up to 55% area savings, thus effectively demonstrating a new avenue for automated generation of approximate accelerators.

1941

We present new high-resolution snow depth data on Arctic sea ice derived from airborne microwave radar measurements from the IceBird campaigns of the Alfred Wegener Institute (AWI) together with a new retrieval method using signal peakiness based on an intercomparison exercise of colocated data at different altitudes. We aim to demonstrate the capabilities and potential improvements of radar data, which were acquired at a lower altitude (200 ft) and slower speed (110 kn) and had a smaller radar footprint size (2-m diameter) than previous airborne snow radar data. So far, AWI Snow Radar data have been derived using a 2–18-GHz ultrawideband frequency-modulated continuous-wave (FMCW) radar in 2017–2019. Our results show that our method in combination with thorough calibration through coherent noise removal and system response deconvolution significantly improves the quality of the radar-derived snow depth data. The validation against a 2-D grid of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">in situ</i> snow depth measurements on level landfast first-year ice indicates a mean bias of only 0.86 cm between radar and ground truth. Comparison between the radar-derived snow depth estimates from different altitudes shows good consistency. We conclude that the AWI Snow Radar aboard the IceBird campaigns is able to measure the snow depth on Arctic sea ice accurately at higher spatial resolution than but consistent with the existing airborne snow radar data of NASA Operation IceBridge. Together with the simultaneous measurements of the total ice thickness and surface freeboard, the IceBird campaign data will be able to describe the whole sea-ice column on regional scales.

In stochastic estimation problems, we aim to estimate an unknown state of interest given a set of measurements received from noisy sensory devices, such as radar, Light Detection and Ranging (LiDAR), etc. A common model of the measurements’ random error is white Gaussian noise. This noise model is usually used to derive an estimator of the unknown state; accordingly, the state is described as a random variable with a Gaussian density. In some applications, the unknown state is known to take only positive values, e.g. estimating size or dimension. In such a case the classical approaches based on Gaussian densities might fail, i.e., produce a negative value. In this work, a recursive Bayesian filter is proposed based on modelling the unknown state as Wishart distributed random matrix. This model ensures that the probability densities of the random variables are restricted to positive real values, even though the measurement’s noise is still modeled as white Gaussian noise. The feasibility of the proposed Bayesian Wishart filter is demonstrated within the framework of Extended Target Tracking (ETT). A target contour measurement model of a symmetric star-convex shape is presented and integrated in the proposed filter to track the target’s extent.

Knowing the distance to nearby objects is crucial for autonomous cars to navigate safely in everyday traffic. In this paper, we investigate monocular depth estimation, which advanced substantially within the last years and is providing increasingly more accurate results while only requiring a single camera image as input. In line with recent work, we use an encoder-decoder structure with so-called packing layers to estimate depth values in a self-supervised fashion. We propose integrating a joint pre-training of semantic segmentation plus depth estimation on a dataset providing semantic labels. By using a separate semantic decoder that is only needed for pre-training, we can keep the network comparatively small. Our extensive experimental evaluation shows that the addition of such pre-training improves the depth estimation performance substantially. Finally, we show that we achieve competitive performance on the KITTI dataset despite using a much smaller and more efficient network.

<strong class="journal-contentHeaderColor">Abstract.</strong> We present a high-resolution airborne radar data set (EGRIP-NOR-2018) for the onset region of the Northeast Greenland Ice Stream (NEGIS). The radar data were acquired in May 2018 with the Alfred Wegener Institute's multichannel ultra-wideband (UWB) radar mounted on the Polar 6 aircraft. Radar profiles cover an area of <span class="inline-formula">∼24</span> 000 km<span class="inline-formula">²</span> and extend over the well-defined shear margins of the NEGIS. The survey area is centered at the location of the drill site of the East Greenland Ice-Core Project (EastGRIP), and several radar lines intersect at this location. The survey layout was designed to (i) map the stratigraphic signature of the shear margins with radar profiles aligned perpendicular to ice flow, (ii) trace the radar stratigraphy along several flow lines, and (iii) provide spatial coverage of ice thickness and basal properties. While we are able to resolve radar reflections in the deep stratigraphy, we cannot fully resolve the steeply inclined reflections at the tightly folded shear margins in the lower part of the ice column. The NEGIS is causing the most significant discrepancies between numerically modeled and observed ice surface velocities. Given the high likelihood of future climate and ocean warming, this extensive data set of new high-resolution radar data in combination with the EastGRIP ice core will be a key contribution to understand the past and future dynamics of the NEGIS. The EGRIP-NOR-2018 radar data products can be obtained from the PANGAEA data publisher (<span class="uri">https://doi.pangaea.de/10.1594/PANGAEA.928569</span>; <span class="cit" id="xref_altparen.1"><a href="#bib1.bibx21">Franke et al.</a>, <a href="#bib1.bibx21">2021</a><a href="#bib1.bibx21">a</a></span>).

Abstract The ice stream geometry and large ice surface velocities at the onset region of the Northeast Greenland Ice Stream (NEGIS) are not yet well reproduced by ice sheet models. The quantification of basal sliding and a parametrization of basal conditions remains a major gap. In this study, we assess the basal conditions of the onset region of the NEGIS in a systematic analysis of airborne ultra‐wideband radar data. We evaluate basal roughness and basal return echoes in the context of the current ice stream geometry and ice surface velocity. We observe a change from a smooth to a rougher bed where the ice stream widens, and a distinct roughness anisotropy, indicating a preferred orientation of subglacial structures. In the upstream region, the excess ice mass flux through the shear margins is evacuated by ice flow acceleration and along‐flow stretching of the ice. At the downstream part, the generally rougher bed topography correlates with a decrease in flow acceleration and lateral variations in ice surface velocity. Together with basal water routing pathways, this hints to two different zones in this part of the NEGIS: the upstream region collecting water, with a reduced basal traction, and downstream, where the ice stream is slowing down and is widening on a rougher bed, with a distribution of basal water toward the shear margins. Our findings support the hypothesis that the NEGIS is strongly interconnected to the subglacial water system in its onset region, but also to the subglacial substrate and morphology.

The comprehension of the behavior of objects in the environment of an automated vehicle is essential information to be extracted out of traffic measurement data. It can be made accessible by applying a data description process consisting of a cluster analysis and subsequent classification step. In this paper, two different algorithms, TRACLUS and Self-Organizing Maps are utilized for a clustering of laser scanner measurement data, recorded by experimental vehicles in real traffic. The results are evaluated and discussed. A remaining open issue of a cluster analysis is the necessity for a classification of clusters, which is missing or solved through manual solutions in related work. We address the matter with a novel approach for an automated classification of clusters using reference trajectories, which is successfully validated for a first exemplary maneuver on highway measurement data.