Virtual Vehicle Validation: Extending the “Cloud Car” Ecosystem

Modern cars have on average, around 100 million lines of code in their software systems. As with hardware components, each software component needs to be thoroughly tested to ensure reliability, safety, and performance.

The problem: Actual in-vehicle testing of software features occurs quite late in the product development lifecycle.  Automakers find themselves identifying software problems at the Start of Production (SOP), and sometimes after the vehicle is launched in the market. That partly explains why, in just the first eight months of 2024, more than 21 million vehicles in the US market were recalled due to 635 separate issues.

As much as automakers might like to perform in-vehicle software tests earlier in the product development cycle, they recognize the challenge: Getting a vehicle environment ready with all the required hardware and testing equipment in early stage of the verification and validation cycle (V-cycle) is almost impossible. Once automakers have a complete prototype of the hardware platform, there is little appetite to delay production simply to fine-tune software.

The most promising solution, from our perspective, is a creative compromise: testing software in a virtual replica of the vehicle platform. To that end, our Cloud Car ecosystem enables the auto industry to reimagine how safety and entertainment features are developed, deployed, and maintained.

Within the larger Cloud Car ecosystem, one key innovation is our Virtual Vehicle Validation (V3P) framework. Vehicle manufacturers now have access to multiple enterprise platforms to create the virtual vehicle environments. What they lack is a unique validation framework that enables them to perform the hardware-in-the-loop (HiL) testing in virtual environments. V3P fills that gap.

While today’s vehicle manufacturers and the vehicle equipment manufacturers understand the needs for thorough software testing using a “shift left” strategy, they continue to perform software testing in the host environment with software-in-the-loop (SIL) test methodology instead of testing the in-vehicle environment following the hardware-in-the-loop (HIL) testing methodology.

Virtual vehicle validation is a promising middle ground, but the challenges of testing in virtual vehicle environments are many, and include:

1. Comprehensive Simulation Requirements: Developing a virtual vehicle environment isn’t just about virtualizing a microcontroller or a System on Chip (SoC) with MPU, GPU, MCU, or ASIC devices. It also involves simulating numerous additional sensors, actuators, and peripheral devices. This complexity can significantly increase the development time and cost, potentially compromising the efficiency and purpose of the virtual platform.

2. High Development Costs and Time: The extensive requirements for accurately simulating all components of a vehicle can lead to substantial costs and time investments. This can be a major barrier, especially in the early stages of development when resources might be limited.

3. Virtualized Hardware-in-the-Loop (HIL) Test Bench: Creating a virtualized HiL test bench presents similar challenges. It requires precise emulation of HIL benches and real-time capabilities to ensure accurate testing. The complexity and cost of setting up such a test bench can be prohibitive.

4. Integration and Compatibility Issues: Ensuring that all virtualized components work seamlessly together can be difficult. Compatibility issues between different virtualized elements can lead to additional debugging and development time.

5. Performance and Real-Time Constraints: Achieving real-time performance in a virtual environment is challenging. The virtual platform must be capable of handling real-time data processing and communication, which can be difficult to achieve without significant computational resources.

6. Validation and Verification: Ensuring that the virtual environment accurately represents the physical vehicle is crucial. This requires extensive validation and verification processes, which can be time-consuming and complex.

The challenges of virtual vehicle validation also extend to original equipment manufacturers (OEMs). To bring OEM components into a virtual vehicle validation platform, vehicle manufacturers and their OEM partners will need to focus on the following priorities:

1. Requirement Decomposition: OEMs typically start with high-level requirements based on the product’s placement and functionalities within the vehicle. As the development progresses, these requirements are broken down into more specialized requirements at various levels, from system to composite to individual components. This decomposition ensures that every aspect of the product is thoroughly defined and understood.

2. Stage-Specific Testing: Given the hierarchical nature of these requirements, it’s essential that the product software is tested against the specified requirements at each stage of development. This ensures that the software meets the necessary criteria at every level, from high-level system requirements to detailed component specifications.

3. Comprehensive Test Environment: To achieve this, a comprehensive test environment is needed. This environment must be capable of:

   • Simulating Different Levels of Abstraction: The test environment should support simulations at various levels, from system-level tests to component-level tests.

   • Traceability: It should provide traceability to ensure that all requirements are covered and tested appropriately. This includes maintaining a clear link between requirements and test cases.

   • Flexibility and Scalability: The environment must be flexible enough to adapt to changing requirements and scalable to handle the complexity of modern automotive systems.

   • Automated Testing: Automation is key to efficiently managing the extensive testing required. Automated test scripts and tools can help ensure consistency and reduce the time and effort needed for testing.

4. Validation and Verification: The test environment should facilitate both validation (ensuring the product meets the needs of the user) and verification (ensuring the product meets the specified requirements). This dual approach helps in identifying issues early and ensures a higher quality product.

By addressing these aspects, OEMs can ensure that their product software is rigorously tested and meets all specified requirements at each stage of development.

Realizing the importance of virtualization, many players now want to enable left-shift. Wipro, as a pioneer in automotive innovation, holds deep understanding in this technology and such toolchain being implemented in the industry, few of which are mentioned below.

1. Continental’s vECU Creator as a part of Continental Automotive Edge (CAEdge) framework, runs on AWS’ cloud service and helps create Virtual ECUs.

2. Virtual Engineering Workbench by Stellantis, Blackberry QNX and AWS offers environments with pre-installed industry-standard tools like Vector AUTOSAR and dSPACE Simulation tools in a supported environment

3. Synopsys Virtualizer Synopsys provides the Synopsys Virtualizer for creating virtual ECUs for full binary software stack validation

4. Vlabworks VLAB VDMs Vlabworks provides VLAB Virtual Device Models (VDMs) for simulating vehicle ECUs, sensors, actuators, and peripheral devices.

5. Siemens PAVE360 Siemens provides the PAVE360 solution for creating virtual ECUs, SoCs, sensors, actuators, and peripheral devices. PAVE360 supports hardware and software co-design, enabling pre-silicon validation, system-level simulations and System of System use cases.

Wipro has the necessary knowledge and skills, we encompass various Test Units as a part of our test framework to deliver virtualization solutions using any tool, to be able to provide be-spoke solutions to our customers.

While third-party vehicle virtualization platforms can enable vehicle manufacturer to create the virtual vehicle environments, they need a unique validation framework that can enable them to perform the HiL testing in virtual environments. In addition to that, vehicle manufacturers need platform agnostic solutions where requirement engineering, validations, and simulations all can be done with high degree of consistency, using the right toolchain. Performing these in silos does not give an integrated picture. Wipro creates an integrated view, that is one of the key success factors for embracing SDV way of development

Wipro’s Virtual Vehicle Validation Platform (V3P) fulfills all of the needs of both vehicle and vehicle equipment manufacturers, and supports all 4 levels of ECU virtualization.By integrating requirement engineering, validation, and simulation into a single, cohesive framework, the platform provides a more holistic and consistent approach to vehicle development. The key features of Wipro V3P include:

1. AI-Driven Test Scripts Generator: This feature automates the creation of test scripts directly from requirements, ensuring that tests are aligned with the specified needs and reducing manual effort.

2. Keyword-Driven Test Framework: Using Behavior-Driven Development Approach, this supports testing at various stages of product development, ensuring that each phase meets the specified requirements.

3. Domain-Specific Test Units: These units are designed to be adaptable to different virtual environments, making the platform versatile and compatible with existing tools like Synopsys Virtualizes, Siemens Pave360, and Vlabworks VLAB VDMs.

By providing a unified validation framework, V3P can facilitate Hardware-in-the-Loop (HIL) testing in virtual environments, which is a significant advancement. This approach not only enhances the transparency and quality of the development process but also helps in reducing costs by minimizing the need for physical hardware.

Wipro’s V3P solution enables simultaneous component development and integration with validation at each checkpoint. Business benefits include:

1. Fail Cheaper and Faster: By adopting different levels of fidelity for verification and validation, OEMs can identify and address issues early, reducing costs and time.

2. Transparency in Product Development: A unified pipeline ensures that all stakeholders have visibility into the development process, enhancing collaboration and accountability.

3. Improved Quality: Extensive verification in various simulated environments helps in identifying potential issues early, leading to a higher quality product.

4. Cost Savings: Reducing the reliance on physical hardware for verification and performing early-stage validation can significantly cut down costs.

The solution approach of V3P aligns well with the SDV paradigm, where development progresses from individual components to a fully integrated system-of-systems. It addresses the unique challenges of SDV development, which lacks the spatial boundaries of traditional ECU-based systems.