Hardware-in-the-Loop alternative approach for an ESP verification 2021-36-0066

With the number and complexity increase of electronic control units (ECUs) within the automotive industry, the tests requirement’s stringency also tends to increase, regarding safety, performance and quality, which in turn lead to greater costs and time required to implement and execute these tests. This work presents an integrated and automated workflow to setup, execute, monitor and evaluate a stress test for an electronic stability program (ESP) module, that may also be applied to other ECUs with minor changes. The proposed method consists on the repeated reproduction of several controller area network (CAN) logs, collected from a prototype, production vehicle or even a virtual automobile, into a real-time bench mounted system, that contains the targeted ECU and peripherals components. A Hardware-in-the-Loop (HiL) approach is used to reproduce physical sensors and CAN inputs to the ECU. The goal of this testing procedure workflow is to allow results comparison between the bench-mounted and in-vehicle or simulated ECUs, on the assumption that the one from which the CAN logs were originally collected, is the reference for correct and well-timed responses. The expected results should be very much alike, considering a proper bench setup, along with the usage of equivalent ECUs and peripherals. By analyzing and comparing the CAN log results, computational failures may be detected, as well as performance issues, in terms of response time. Some challenges arise for each ECU specific’s actuation methods and interactions with other electronic systems, concerning physical setup and integration solutions. Regarding the addressed ESP unit, some sensors required a specific instrumentation level in order to simulate physical inputs to the module, for instance, wheel speed sensors. The ESP was successfully integrated to the real-time bench and replied to the physical and CAN stimuli appropriately, as if it were properly assembled and connected to a vehicle’s network, therefore this study case is a proof of concept for the proposed workflow.