In CCAM, system architectures span local computing nodes, in-car heterogeneous systems connected via wired networks, and cloud-connected vehicles and infrastructure via wireless networks. This objective aims to provide HAL interfaces for such multilevel architectures, enabling seamless integration, hardware-agnostic software development, and a time/resource-composable System-on-Module (SoM) platform. At the device level, the HAL will offer HW-independent primitives to configure and monitor communication and computing nodes (e.g., high-performance COTS devices), supporting higher-level SASE requirements. It will also include configurations for safe, secure operation and updates across the distributed system, ensuring flexibility, portability, and consistent SASE through hardware abstraction.

Enable platform-independent development, validation, deployment, and seamless integration of application software across diverse CCAM cloud-edge architectures. To support dynamic updates and car-to-cloud/cloud-to-car task reallocation, a unified software abstraction layer (OAL) is needed above the HAL to isolate application software from hardware and low-level dependencies. Built on open-source hypervisors, OS, and middleware, the OAL will provide a common interface for applications both on vehicles and in the cloud. Container technologies and adaptable communication middleware will ensure transparent execution and communication across nodes and with the cloud.

A key enabler for rolling out Level 4+ automated driving is the ability for CCAM systems to evolve in operation—supporting SW updates, HW upgrades, and dynamic task shifting between vehicles and the cloud. This objective targets a dynamic update and orchestration runtime platform (ORP) that enables safe, secure, and synchronized over-the-air (SOTA) updates, HW upgrades, and vehicle-to-cloud (V2C) task reallocation without requiring service-station visits. It involves defining the necessary processes, orchestration architecture, and safety/security concepts based on the OAL and HAL. Integration of local and remote verification services ensures compatibility and system validity. Ultimately, this will accelerate Level 4+ deployment by connecting cloud-based development with real-world fleet operation.

Develop a model-based engineering (MBE) methodology and toolchain tailored for next-gen AI-driven and sensor-fused automated vehicle applications. Grounded in virtualization, this toolchain will streamline and automate the design, implementation, and validation of software modules, ensuring compliance with safety regulations and quality standards while reducing human error. It will include an open toolchain for modeling and automated code generation to produce reliable, certifiable code for the SASE UP2DATE4SDV architecture based on HPC COTS. To accelerate SDV development and ensure compliance, a harmonized validation and verification framework will be provided, leveraging virtualization of the upgradable in-vehicle control architecture.

This objective defines a reference zonal vehicle architecture to support SASE in the creation of a cloud-connected, reconfigurable blueprint (O1) and a resource-composable software deployment and execution layer (O2). It also introduces a standardized Combined Safety-Security Update Management and Execution Process (O3) to enable dynamic updates, orchestration, and reconfiguration while maintaining SASE. Reactive and preventive mechanisms will address safety and cybersecurity risks arising from the complex, connected UP2DATE4SDV architecture and AI-based applications. All concepts will be evaluated by external certification bodies to ensure compliance with safety, security, and update-related regulations.

The project will develop six Use Cases (UCs) to evaluate the outcomes of O1–O5 and demonstrate the value of the proposed concepts for highly connected and automated vehicles. These demonstrations will highlight how the solutions align with CCAM priorities: centralized, reliable, cyber-secure, upgradable, and connected. The initial UCs focus on core technologies—UC1: HW/SW updates in the perception chain, UC2: reliability, UC3: hardware resource constraints, and UC4: cybersecurity. The final UCs (UC5 and UC6) integrate these technologies in lab environments using HIL and driving simulators to showcase advanced scenarios like perception stack upgrades and connected automated driving.