Within this paper we present a new concept on deriving test cases from simulation data and outline challenging tasks when testing and validating fully automated driving functions and Advanced Driver Assistance Systems (ADAS). Open questions on topics like virtual simulation and identification of relevant situations for consistent testing of fully automated vehicles are given. Well known criticality metrics are assessed and discussed with regard to their potential to test fully automated vehicles and ADAS. Upon our knowledge most of them are not applicable to identify relevant traffic situations which are of importance for fully automated driving and ADAS. To overcome this limitation, we present a concept including filtering and rating of potentially relevant situations. Identified situations are described in a formal, abstract and human readable way. Finally, a situation catalogue is built up and linked to system requirements to derive test cases using a Domain Specific Language (DSL).
Radar signals have been dramatically increasing in complexity, limiting the source separation ability of traditional approaches. In this paper we propose a Deep Learning-based clustering method, which encodes concurrent signals into images, and, for the first time, tackles clustering with image segmentation. Novel loss functions are introduced to optimize a Neural Network to separate the input pulses into pure and non-fragmented clusters. Outperforming a variety of baselines, the proposed approach is capable of clustering inputs directly with a Neural Network, in an end-to-end fashion.
We describe a new wave-front sensor based on the efficient Shack-Hartmann quad cell system. The key improvement to existing designs is a modified lenslet array. The traditional refractive lenslet array is replaced with a segmented reflective system that allows individual control of each subaperture. This system has produced diffraction-limited slope correction of the six 1.83-m mirrors of the MMT using & 9.9V magnitude guide star and significant image improvement on guide stars as faint as 11.3V magnitude, limited by the readout noise of the CCD quad cell detector. Recent experiments with an improved detector indicate diffraction-limited imaging on guide stars as faint as the 15th magnitude. This optimized wave-front sensor, equally applicable to filled-aperture telescopes, promises to extend the amount of sky coverage available for adaptive correction in the near IR.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.