Efficient Neural Network Implementations on Parallel Embedded Platforms Applied to Real-Time Torque-Vectoring Optimization Using Predictions for Multi-Motor Electric Vehicles

The combination of machine learning and heterogeneous embedded platforms enables new potential for developing sophisticated control concepts which are applicable to the field of vehicle dynamics and ADAS. This interdisciplinary work provides enabler solutions -ultimately implementing fast predictions using neural networks (NNs) on field programmable gate arrays (FPGAs) and graphical processing units (GPUs)- while applying them to a challenging application: Torque Vectoring on a multi-electric-motor vehicle for enhanced vehicle dynamics. The foundation motivating this work is provided by discussing multiple domains of the technological context as well as the constraints related to the automotive field, which contrast with the attractiveness of exploiting the capabilities of new embedded platforms to apply advanced control algorithms for complex control problems. In this particular case we target enhanced vehicle dynamics on a multi-motor electric vehicle benefiting from the greater degrees of freedom and controllability offered by such powertrains. Considering the constraints of the application and the implications of the selected multivariable optimization challenge, we propose a NN to provide batch predictions for real-time optimization. This leads to the major contribution of this work: efficient NN implementations on two intrinsically parallel embedded platforms, a GPU and a FPGA, following an analysis of theoretical and practical implications of their different operating paradigms, in order to efficiently harness their computing potential while gaining insight into their peculiarities. The achieved results exceed the expectations and additionally provide a representative illustration of the strengths and weaknesses of each kind of platform. Consequently, having shown the applicability of the proposed solutions, this work contributes valuable enablers also for further developments following similar fundamental principles.

Development of a Model of on-Board Pemfc Powered Locomotive with a Metal Hydride Cylinder

ABSTRACTThis paper presents a phase-zero evaluation case of installing on-off-board hybrid powered Electric Motor Vehicle (EMV) in existing and new local line and reports development of a model fuel cell powered locomotive. EMV such as electric car and locomotive are a new conceptual EMV using hybrid power between off-board substation and on-board Regenerative Fuel Cell (RFC) power system with Metal Hydride (MH) stored hydrogen generated with water electrolyzer by off-board surplus power. In this study, it is estimated a possibility to close power gap over 30 % in placing the new conceptual vehicle. The Locomotive is a 110 cm long locomotive powered by a 20 W PEMFC configured with 20 cells and supplied with about 2 g hydrogen, from a cylinder of 100 g metal hydride, and natural convection air (02). Measuring 50 cm (W), 50 cm (H), and weighting 25.9 kgf, the locomotive has a permanent magnet motor with a rated power 38 W (12 V, 3 A) and ran on railway that has a gauge of 126 mm (3'6“/8.5=4.94”), a length of 100 m. The performance of this train was acceleration of 0.5 m/s, cruising speed of 4.1 m/s at traction force of 15.8 N (1.6 kgf), average rolling friction of 5 N (460 gf).