Intense competition and technological evolution demand a new approach to accelerate vehicle function development . By Megan Lampinen
The rise of the software-defined vehicle (SDV) and electrification are reshaping how vehicles are developed. System complexity is growing exponentially: some of today’s vehicles feature more than 100 million lines of code, 150 ECUs and 3,000 semiconductors. This is further complicated by the transition from combustion engines (ICE) to hybrid and battery-electric powertrains. Calibrating thousands of signals across integrated multi-ECU systems for optimal range, performance, and lifespan is an immense task.
“EVs bring different challenges around range, energy efficiency and thermal management, but hybrids bring all of that plus the complexity of the ICE emissions calibration,” says Gerit von Schwertführer, Vice President Solution Field Integrated Calibration and Application (ICA) at ETAS, a 100% Bosch subsidiary. “It’s clear that teams cannot manage the setup and calibration of these systems using traditional ways of working.”
While complexity is growing, development cycles need to shrink. Incumbent automakers have become accustomed to launching a new vehicle model every six to eight years, with a facelift at some point in between. The rise of new Chinese players, with their software-centric approaches, has lifted the bar. “Some Chinese OEMs are talking about development times of 12 to 15 months—from the very start of a project to the start of production,” notes von Schwertführer. “The market gets used to this innovation speed. Europe might be okay waiting four years for a facelift but try that in China and you are out of the game.”
Vehicle complexity is increasing
Speed to market is particularly important for EVs, where battery and charging technology are developing so rapidly that today’s state-of-the art quickly becomes outdated. Faster working is also a major cost-saver: “If you are twice as fast in bringing new vehicles to market, you can be sure it will also be much cheaper. You simply don’t have the time to burn as much money as if you had twice the development period.”
Mastering multi-ECU calibration
Von Schwertführer believes the key to success in this complex and evolving landscape is to break up siloes and introduce more automation and AI into the development process. A software-centric development paradigm is a good place to start. “Without it, you end up with an exponential increase in complexity,” he tells Automotive World. “That means development times increase and/or your team size increases, neither of which are valid options.”
Achieving this software-centric vision, however, is intrinsically linked to the ability to precisely measure and calibrate across the entire system. Without accurate data and the means to fine-tune parameters, even the most innovative software architecture can falter. This is where the critical role of robust measurement and calibration tools becomes undeniable, forming the very backbone of efficient and effective multi-ECU development.
ETAS is helping developers tackle speed challenges and system complexity with a range of ECU access technology solutions. “Mastering multi-ECU calibration optimally is one of the biggest challenges. It starts with systems that can measure each and every ECU, laying the foundation for measurement and calibration access to all the levers you need to have under control.” The company’s ETK (or Extended Testing and Calibration) serves as a universal ECU interface for development applications within engine and transmission ECUs.
After laying that physical foundation, teams need access to synchronised data from the ECUs and sensors. “You need a central steering software that provides access to all these data flows in a simple way. You simply see the data flow, the parameters you can change, and the effect,” he says.
Our tools, especially INCA, are perfectly suited to support a development cycle of 12 to 15 months
With this in place, there is then the option to add automation and AI-support. “You can calculate system models with machine learning using measurement data and easily find all the dependencies and ultimately the sweet spot, even in between different local optima.”
INCA (Integrated Calibration and Application Tool) is ETAS’ central tool for calibration tasks. It currently has more than 50,000 users worldwide and comes with an ecosystem of adjacent tools, such as INCA-Flow for the automation of calibration processes and ASCMO (Advanced Simulation for Calibration, Modelling and Optimisation) for AI-supported optimum calibration setups. This suite of solutions is used by automakers, Tier 1s, and engineering service providers. “There are few Tier 1s not working with INCA,” adds von Schwertführer.
The impact
The impact of using such tools could be significant. Von Schwertführer estimates that ETAS can improve development efficiency by up to 30%, emphasising that every use case varies and results depend heavily on existing work practices. Regardless of the specific figure, the overall efficiency gains open the door to faster development. “This means that our tools, especially INCA, are perfectly suited to support a development cycle of 12 to 15 months,” he asserts.
That’s good news for incumbents: the market is full of EV hopefuls, with more than 100 in China alone. As battery technology matures, it will become increasingly difficult to stand out and differentiation will need to focus on more than just range and charging time. “If everyone is competing on these areas, you are most likely not the leader,” warns von Schwertführer. Many EV brands are already seeking to differentiate by vehicle design or IVI features. The next battleground could be vehicle dynamics.
He highlights BMW’s recently unveiled iX3, the first model of the Neue Klasse, as an example of what is possible with optimised steering, braking, suspension and powertrain. “When you find the overall optimum, you have a unique and specific driving behaviour that sets you apart from others. OEMs will increasingly look to differentiate on this aspect and design their vehicle dynamics to support brand values.”
Whether it’s vehicle dynamics or electrified powertrains, an AI-supported software-centric development approach will go far in securing players a competitive advantage, but it requires a complete rethink of historic strategies. That comes with risk. “Taking risks is simply part of the Chinese game,” concludes von Schwertführer. “The big question is to what extent risk can be accepted and established in safety-driven markets like Western Europe and the US.”
