Saab Automobile AB

About: Saab Automobile AB is a based out in . It is known for research contribution in the topics: Radar & Antenna (radio). The organization has 760 authors who have published 890 publications receiving 11811 citations.

Instructor Based Training versus Computer Based Training—A Comparative Study

Abstract: This article describes two studies conducted to compare assembly performance and learning rate between computer based training and traditional training of skilled assembly operators. The studies we ...

True-Time-Delay Beamforming Receiver With RF Re-Sampling

Abstract: Analog domain true-time-delays (TTD) are desired in hybrid beamforming receivers with large relative bandwidths to mitigate the problem of beam squint. We propose a true-time-delay beamforming receiver architecture which enables squint-free wideband spatial filtering prior to the A/D conversion. The receiver implements true-time-delay with delayed re-sampling of the discrete-time output of a passive mixer. The receiver has the capability to extend the range of the beamforming delays from one to several carrier periods of the RF signal with pulse-skipped local oscillator (LO) signals, thereby enabling TTD beamforming with large antenna arrays. Further, a polyphase structure with parallel mixers is proposed to prevent spectral aliasing resulting from the lowered sample rate of the pulse-skipped LO signals. In addition, the maximum beamforming delay scales with the LO frequency, supporting large arrays also at low frequencies where the antenna separation set by the wavelength is large. We verify the proposed concepts with transistor-level simulation of the receiver implemented with a 28-nm CMOS process. The design achieves a squint-free beamforming for a 400 MHz RF bandwidth, and a maximum beamforming delay of three carrier time periods. The power consumption for a 3 GHz carrier frequency is 4 mW per antenna.

Object Recognition in Forward Looking Sonar Images using Transfer Learning

Abstract: Forward Looking Sonars (FLS) are a typical choice of sonar for autonomous underwater vehicles. They are most often the main sensor for obstacle avoidance and can be used for monitoring, homing, following and docking as well. Those tasks require discrimination between noise and various classes of objects in the sonar images. Robust recognition of sonar data still remains a problem, but if solved it would enable more autonomy for underwater vehicles providing more reliable information about the surroundings to aid decision making. Recent advances in image recognition using Deep Learning methods have been rapid. While image recognition with Deep Learning is known to require large amounts of labeled data, there are data-efficient learning methods using generic features learned by a network pre-trained on data from a different domain. This enables us to work with much smaller domain-specific datasets, making the method interesting to explore for sonar object recognition with limited amounts of training data. We have developed a Convolutional Neural Network (CNN) based classifier for FLS-images and compared its performance to classification using classical methods and hand-crafted features.

Hydraulic pump for a vehicle engine

Abstract: A hydraulic pump in the form of an oil pump (12) for a vehicle engine has both its pinion (21) which is driven by the engine's crankshaft (3), and a ring-gear (22) which cooperates with the pinion (21), incorporated in an end-cover (11) fastened to the end of the engine block (2). A cap (15) is inserted in a collar (14) on the side of the end-cover facing away from the engine and is only held in position there by means of a locking ring (17) which engages with a groove (16) in the collar and presses the cap (15) towards the engine. Outside the oil pump, a partly socket-shaped belt pulley (7) is mounted on the crankshaft (3) and its socket-shaped portion substantially grips the collar. Fitting and removal are thus simplified and engine overall length reduced. Simplified sealing also reduces oil leakage risks.

Architectural Challenges in Memory-Intensive, Real-Time Image Forming

Abstract: The real-time image forming in future, high-end synthetic aperture radar systems is an example of an application that puts new demands on computer architectures. The initial question is whether it is at all possible to meet the demands with state-of-the-art technology or foreseeable new technology. It is therefore crucial to understand the computational flow, with its associated memory, bandwidth and processing demands. In this paper we analyse the application in order to, primarily, understand the algorithms and identify the challenges they present on a basic architectural level. The processing in the radar system is characterized by working on huge data sets, having complex memory access patterns, and doing real-time compensations for flight path errors. We propose algorithm solutions and execution schemes in interplay with a two-level (coarse-grain/fine-grain) system parallelization approach, and we provide approximate models on which the demands are quantified. In particular, we consider the choice of method for the performance- intensive data interpolations. This choice presents a trade-off problem between computational performance and size of working memory. The results of this "upstream " study will serve as a basis for further, more detailed architecture studies.