An Intelligent System-on-a-Chip for a Real-Time Assessment of Fuel Consumption to Promote Eco-Driving

TL;DR

The paper presents a SOM-based intelligent system for real-time eco-driving advice, implemented on a FPGA-enabled PSoC to classify individual driving styles and suggest personalized control adjustments. It leverages real-world instrumented-car data (Uyanik dataset) and GT-Suite fuel simulations to train and label SOM clusters, achieving up to $9.5\%$ to $31.5\%$ potential fuel/emission reductions with higher gains for engaged users. The HW/SW co-design features a fully parallel SOM accelerator on the FPGA with $M=121$ neurons and $N=4$ features, delivering a latency of 12 cycles at up to ~129.7 MHz and ~0.3 W power, while the software part provides I/O, windowing, and natural-language advice. The approach demonstrates significant potential for on-board eco-driving assistance and offers a basis for extending to more driving contexts and deployment in actual vehicles.

Abstract

Pollution that originates from automobiles is a concern in the current world, not only because of global warming, but also due to the harmful effects on people's health and lives. Despite regulations on exhaust gas emissions being applied, minimizing unsuitable driving habits that cause elevated fuel consumption and emissions would achieve further reductions. For that reason, this work proposes a self-organized map (SOM)-based intelligent system in order to provide drivers with eco-driving-intended driving style (DS) recommendations. The development of the DS advisor uses driving data from the Uyanik instrumented car. The system classifies drivers regarding the underlying causes of non-optimal DSs from the eco-driving viewpoint. When compared with other solutions, the main advantage of this approach is the personalization of the recommendations that are provided to motorists, comprising the handling of the pedals and the gearbox, with potential improvements in both fuel consumption and emissions ranging from the 9.5\% to the 31.5\%, or even higher for drivers that are strongly engaged with the system. It was successfully implemented using a field-programmable gate array (FPGA) device of the Xilinx ZynQ programmable system-on-a-chip (PSoC) family. This SOM-based system allows for real-time implementation, state-of-the-art timing performances, and low power consumption, which are suitable for developing advanced driving assistance systems (ADASs).

Figures (17)

• Figure S1: Data-acquisition systems and sensors installed in the Uyanik car abut2009real.
• Figure S2: Offline sequence of tasks involved in the design and development of an self-organized map (SOM)-based intelligent system for fuel consumption assessment. The dotted arrows indicate that the simulated fuel consumption data are also used to label the SOM-based clustering for verification purposes and to elaborate the hardware (HW) implementation of the SOM.
• Figure S3: Flow of real-world telemetry-based fuel consumption simulation. It has macroscopic car parameters (gear ratios, tyre dimensions, and wheelbase) and telemetry (gas pedal, brake pedal, speed, selected gear, and accelerations) as inputs. The model returns the simulated fuel flow as output.
• Figure S4: Block diagram of real-world telemetry-based fuel consumption simulation of Uyanik Renault Mégane 1.5 dCi Sedan 74 kW. It has macroscopic car parameters (gear ratios, tyre dimensions, and wheelbase) and telemetry (gas pedal, brake pedal, speed, selected gear, and accelerations) as inputs. The model returns the simulated fuel flow as output.
• Figure S5: Comparison of measured data vs. simulation results. (a) The inferred gear considering the computed rpm/speed ratios. (b) The measured revolutions per minute (RPM) of the vehicle vs. the RPM simulated by the model. (c) The measured speed of the vehicle vs. the speed simulated by the model.