How to optimize your ECU software development process

As E/E architectures move to a more centralized model, microcontroller-based ECUs will remain essential. However, development cycles must be accelerated to meet the demands of the functionality-driven market, especially with the advent of software-defined vehicles.

Despite this shift, the V-model remains a relevant framework for embedded software development, effectively combining established processes with innovative solutions. The development process is optimized along the key work steps. These do not necessarily follow a strict sequence; some can occur in parallel or be repeated, and requirement and variant management are considered throughout.

In an ECU, a microcontroller manages specific interlinked functions such as engine and vehicle dynamics control. A detailed requirements catalog is essential, starting with identifying functionalities, constraints, hardware elements, middleware, and their interdependencies. All components must comply with strict automotive standards from the outset, which requires careful attention to detail.

The vehicle software platform bridges the hardware or operating system and the application software in an ECU, abstracting hardware and offering a standardized interface for smooth interaction among software components. Such an intermediary must adhere to the highest security standards, as vulnerabilities in the ECU can result in unauthorized access and data breaches, compromising vehicle safety. The selection of the vehicle software platform must emphasize maturity (ISO 26262 ASIL-D compliance), future-proofing (no vendor lock-in), and cybersecurity. A robust, configurable solution with continuous updates is essential to protect the ECU from evolving threats and meet regulatory requirements.

After defining the architecture and identifying component requirements, developers translate the system design into functional software. They create new functions, fix errors, and optimize existing functionss. The challenge is to establish a process that allows for implementing exixting functions without rewriting them, while maintaining functional safety, cybersecurity, and code efficiency.

At this stage, the vehicle software platform configuration, architecture design, application software, and pre-calibration data are compiled into code for the microcontroller. One of the key challenges is to match the performance of the vehicle function to the capabilities of the new hardware, while meeting efficiency goals and environmental conditions. This requires fine-tuning the vehicle software platform for resource utilization and real-time operation, as well as rigorous testing to meet safety and regulatory standards.

The testing and validation phase ensures that ECU functions meet safety, performance, and reliability requirements. Extensive testing identifies potential issues, which can be complex and lengthy for software-heavy vehicles. Thus, conducting as many tests as possible in a short time is critical to minimize costs and maximize performance.

During calibration, parameters are populated with data to match software behavior to the physical system. Some applications have thousands of interrelated parameters that must meet performance goals, environmental conditions, and regulatory standards. Adjustments may be required for future updates such as when emissions norms change.

Once the software reaches the “ready for use” stage, it undergoes a final approval process and is fully flashed onto the target ECUs.

For more information on how to optimize your development processes for ECU software, visit our website and read our white paper, Five major challenges in software development for automotive microcontrollers.