%0 Journal Article %1 ARNDT2019113 %A Arndt, Oliver Jakob %A Rallapalli, Parwesh %A Blume, Holger %B Big Data Analytics for Cyber-Physical Systems %D 2019 %E Dartmann, Guido %E Song, Houbing %E Schmeink, Anke %I Elsevier %K 2019 myown %P 113 - 143 %R https://doi.org/10.1016/B978-0-12-816637-6.00006-3 %T Portable implementations for heterogeneous hardware platforms in autonomous driving systems %U http://www.sciencedirect.com/science/article/pii/B9780128166376000063 %X Complex driver-assistance systems that analyze driving situations based on a range of sensors enable autonomous driving vehicles—a key aspect of smart cities. This massive automation necessitates computationally powerful and energy-efficient hardware devices available in each individual driving unit. Heterogeneous multiprocessor system-on-chips provide excellent performance-to-power characteristics for the use in driver-assistance applications. Since these programmable chips use flexible software, they theoretically feature high maintainability and portability. However, due to the lack of programmability of different parallel and heterogeneous processing units, developers can barely fully exploit all computational capabilities. To overcome the gap between theoretical peak performance and the effectively gained speedup, diverse programming approaches and supportive tools have emerged. This work presents an overview of the most important trends and contributes a middleware approach for abstracting, and thus unifying, the programming for homogeneous and heterogeneous architectures. %@ 978-0-12-816637-6

@article{ARNDT2019113, abstract = {Complex driver-assistance systems that analyze driving situations based on a range of sensors enable autonomous driving vehicles—a key aspect of smart cities. This massive automation necessitates computationally powerful and energy-efficient hardware devices available in each individual driving unit. Heterogeneous multiprocessor system-on-chips provide excellent performance-to-power characteristics for the use in driver-assistance applications. Since these programmable chips use flexible software, they theoretically feature high maintainability and portability. However, due to the lack of programmability of different parallel and heterogeneous processing units, developers can barely fully exploit all computational capabilities. To overcome the gap between theoretical peak performance and the effectively gained speedup, diverse programming approaches and supportive tools have emerged. This work presents an overview of the most important trends and contributes a middleware approach for abstracting, and thus unifying, the programming for homogeneous and heterogeneous architectures.}, added-at = {2020-01-09T13:38:31.000+0100}, author = {Arndt, Oliver Jakob and Rallapalli, Parwesh and Blume, Holger}, biburl = {https://www.bibsonomy.org/bibtex/299222ea24eb2119c129c3e638f9bb448/imsl3s}, booktitle = {Big Data Analytics for Cyber-Physical Systems}, description = {Portable implementations for heterogeneous hardware platforms in autonomous driving systems - ScienceDirect}, doi = {https://doi.org/10.1016/B978-0-12-816637-6.00006-3}, editor = {Dartmann, Guido and Song, Houbing and Schmeink, Anke}, interhash = {3ef842c42607004a19d60c7e3bccff23}, intrahash = {99222ea24eb2119c129c3e638f9bb448}, isbn = {978-0-12-816637-6}, keywords = {2019 myown}, pages = {113 - 143}, publisher = {Elsevier}, timestamp = {2020-01-09T13:38:31.000+0100}, title = {Portable implementations for heterogeneous hardware platforms in autonomous driving systems}, url = {http://www.sciencedirect.com/science/article/pii/B9780128166376000063}, year = 2019 }