Abstract

© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Advanced Driver Assistance Systems (ADAS) and Autonomous Driving (AD) have drastically increased the performance demands of automotive systems. Suitable highperformance platforms building upon Graphic Processing Units (GPUs) have been developed to respond to this demand, being NVIDIA Jetson TX2 a relevant representative. However, whether high-performance GPU configurations are appropriate for automotive setups remains as an open question. This paper aims at providing light on this question by modelling an automotive GPU (Jetson TX2), analyzing its microarchitectural parameters against relevant benchmarks, and identifying specific configurations able to meaningfully increase performance within similar cost envelopes, or to decrease costs preserving original performance levels. Overall, our analysis opens the door to the optimization of automotive GPUs for further system efficiency. Peer Reviewed

Abstract

High performance computing (HPC), both at hardware and software level, has demonstrated significant improve- ments in processing large datasets in a timely manner. However, HPC in the field of air traffic management (ATM) can be much more than only a time reducing tool. It could also be used to build an ATM simulator in which distributed scenarios where decentralized elements (airspace users) interact through a centralized manager in order to generate a trajectory-optimized conflict-free scenario. In this work, we introduce an early prototype of an ATM simulator, focusing on air traffic flow management at strategic, pre-tactical and tactical levels, which allows the calculation of safety and efficiency indicators for optimized trajectories, both at individual and network level. The software architecture of the simulator, relying on a HPC cluster of computers, has been preliminary tested with a set of flights whose trajectory vertical profiles have been optimized according to two different concepts of operations: conventional cruise operations (i.e. flying at constant altitudes and according to the flight levels scheme rules) and continuous climb cruise operations (i.e., optimizing the trajectories with no vertical constraints). The novel ATM simulator has been tested to show preliminary benchmarking results between these two concepts of operations. The simulator here presented can contribute as a testbed to evaluate the potential benefits of future Trajectory Based Operations and to understand the complex relationships among the different ATM key performance areas Peer Reviewed