The performance of automotive radar systems is expected to significantly increase in the near future. With enhanced resolution capabilities more accurate and denser point clouds traffic participants roadside infrastructure can be acquired so amount gathered information growing drastically. One main driver for this development global trend towards self-driving cars, which all rely on precise fine-grained sensor information. New signal processing concepts have developed order provide...

Autonomous driving will significantly shape the near future of transportation that requires distinct knowledge environment, especially lane boundaries ahead vehicle. For this sensing task, optical and automotive radar sensors are mostly applied, while sensor is less sensitive to non-ideal illumination conditions. Road can be detected by through objects like guardrails, delineators, road curbs, so on. However, in a multi-lane roadway or with missing roadside infrastructure, adjacent lanes...

Novel automotive high resolution radar sensors can detect several thousands of reflection points from the surrounding environment, e.g., pedestrians, cyclists, vehicles and roadside infrastructure. For object classification tracking, detection belonging to same shall be clustered into one group before further processing. This paper presents an adaptive clustering approach based on a range/angle/velocity-grid generated originally signal processing angle estimation stage. In contrast...

On the way to Highly Automated Driving (HAD), new conditions for sensors used in vehicles arise. To achieve a highly accurate environmental perception resolution of radar has be increased. Only with fine grained sensor information, it is possible maneuver safely and automated at all road urban rural surroundings. A prototypical implementation such sensor, which fulfills these requirements, presented following as automotive 4D radar. The relevant parts baseband signal processing are explained...

Pushed by the EuroNCAP regulations number of autonomous emergency braking systems for pedestrians (AEB-P) is rapidly increasing since year 2016. The same rise expected cyclist protection driven new test scenarios from 2018 onwards. To get an adequate reaction system, target objects have to be classified clearly. Visual sensors provide some benefits in object recognition, but performance suffers adverse environmental conditions. In this area radar proven as very robust against darkness, fog,...

Future high-resolution radars enable new functionalities in advanced driver assistance systems, such as estimation of contour, position, and orientation vehicles on the road. However, straightforward approaches like that Oriented Bounding Box generally fail challenging automotive scenarios, when only one side vehicle is visible to radar. In this paper, an approach based Generalized Hough Transform matching presented examined for its use scenarios. An optimization method discussed finally,...

Clustering of measurement data is an important task in digital signal processing. Especially the case radar processing need clustering detection points becomes obvious when high-resolution sensor systems are used. usually used as a preprocessing step for classification measured data. In this paper new approach automotive presented. A shape finding technique from image processing, called border following, to perform task. Some adjustments and modifications method required get it working with...

Future high-resolution radars enable new functionalities in advanced driver assistance systems, relying on fast and reliable extraction of properties vehicles the road. A critical property for prediction trajectories assessment potentially dangerous situations is that actual motion - velocity vector yaw rate observed objects. In this paper, an approach to distinguish linear from non-linear motions as well estimating using only a single radar sensor presented evaluated via measurements.

Automated vehicles need to detect and classify objects traffic participants accurately. Reliable object classification using automotive radar sensors has proved be challenging. We propose a method that combines classical signal processing Deep Learning algorithms. The range-azimuth information on the reflection level is used extract sparse region of interest from range-Doppler spectrum. This as input neural network (NN) classifies different types stationary moving objects. present hybrid...

Der vorliegende Beitrag untersucht die Wirkung von unternehmerischen Work-Life Balance Initiativen für emotionale Bindung der Mitarbeiter zum Unternehmen, d. h. das organisationale Commitment Mitarbeitern im Kontext Unternehmensberatung. Dazu wurde eine Untersuchung mit insgesamt 275 Unternehmensberaterinnen und -beratern durchgeführt. Die Datenanalyse erfolgte anhand eines PLS-Strukturgleichungsmodells (Partial-Least-Squares-Ansatz). Ergebnisse weisen darauf hin, dass auch in...

Lane Change Assist, Departure Warning, Keeping System, and Automated Driving all require the sensors in vehicles to be able detect road lane boundaries reliably accurately under circumstances. However, such functions rely currently solely on camera sensors, which result unavoidably affects stability of whole system certain extreme lighting conditions. Therefore, this paper shows a verification test novel approach boundary detection with an automotive radar specific kind reflectors. A new...

Prediction of the object movement from sensor data in automotive sector is a widespread research and development topic. Dependent on used types, tracking has been established over several measurement cycles. A prominent example this Kalman filter. In time critical scenarios with less reaction number cycles not suitable. To detect within one single cycle only radar candidate, due to ability measure velocity objects instantaneously by using Doppler effect.A new approach estimate direction...