MOSA.ic.AI Webinar Series

AI/ML is foreseen to be integrated into safety- and mission-critical aerospace and defense systems. Engineering teams now face a hard question: how do you develop, deploy, and verify AI-enabled software with the predictability, traceability, and evidence that certification requires? Lynx and Ansys, part of Synopsys have collaborated to showcase a continuous, closed-loop engineering workflow that links system requirements, AI/ML development, embedded deployment, and system-level verification in an example blueprint built around a Helicopter Emergency Stopping (HES) use case.

On the Ansys side, the demonstration follows a Digital Engineering approach and is based upon a variety of tools to join MBSE (Model Based Systems Engineering) and Simulation: Software Architect Modeler (SAM), medini analyze, SCADE Suite, AVxcelerate Autonomy, AVxcelerate Sensors and Systems Tool Kit. Lynx provides the deterministic embedded execution environment with MOSA.ic.AI, a platform for running AI workloads on predictable CPU/GPU runtime environments designed for mission-critical operation. MOSA.ic.AI provides secure partitioning for mixed-criticality architectures and transforms AI/ML models trained anywhere, to execute bounded and deterministically in this environment. Together, these solutions support system architecture modeling, safety analysis, scenario-based simulation, AI/ML design, development and deployment, model-based software engineering and code generation, and requirements verification. The engineering workflow enables rapid prototyping, ML-model re-training and turnaround, moving from software-in-the-loop simulation to target-based and hardware-in-the-loop verification. 

Attendees will learn how to:

• Use simulation to discover AI-related system underperformance early, before implementation reach target hardware.
• Maintain end-to-end traceability from requirements through AI/ML lifecycle artifacts to verified behavior.
• Integrate deterministic control software and AI/ML functions in a unified, modular, mixed-criticality embedded architecture.
• Transition from SiL to HiL testing in a predictable embedded runtime environment.
• Create and capture safety evidence throughout the workflow to support assurance and certification.

AI innovation is accelerating rapidly, but deploying trained models into mission-critical embedded systems remains a significant challenge. While today's AI frameworks excel at model development, they were not designed for the deterministic inference, workload isolation, and certification requirements of aerospace, defense, and other safety-critical applications. LYNX MOSA.ic.AI bridges this AI deployment gap by optimizing AI inference for Size, Weight, and Power (SWaP)-constrained embedded systems, enabling validated models to execute predictably alongside real-time and safety-critical workloads.

In this webinar, you'll learn how to bring your own trained AI model into MOSA.ic.AI without changing your existing AI development workflow. We'll demonstrate how the platform separates model development from deployment while providing deterministic inference, unified CPU/GPU execution, mixed-criticality support, hardware portability, and open systems architecture to simplify integration and enable reliable AI deployment at the edge.

Attendees will learn how to:

• Bring existing AI models into MOSA.ic.AI without modifying established development workflows.
• Optimize AI inference for SWaP-constrained embedded systems while maintaining predictable, deterministic behavior.
• Deploy AI alongside real-time and safety-critical applications using a unified CPU/GPU architecture.
• Reduce integration risk with an open, hardware-portable platform designed for long-term mission evolution
  
  

Read the Webinar Q&A Blog >

AI workloads introduce timing variability that can make them difficult to deploy in mission- and safety-critical environments. Dynamic resource demands, shared CPU and GPU resources, and unpredictable execution times can interfere with real-time applications and complicate system validation. To deploy AI confidently at the edge, system architects need a way to control how AI workloads execute and demonstrate that critical timing requirements will be met.

In this webinar, we’ll explore how LYNX MOSA.ic.AI enables predictable, bounded AI inference within mixed-criticality embedded systems. Attendees will learn how workload isolation, controlled resource allocation, and unified CPU/GPU execution help reduce interference and establish consistent runtime behavior. We’ll also examine how these capabilities support system validation, certification objectives, and the long-term evolution of AI-enabled platforms.

Attendees will learn how to:

• Identify common sources of variability in embedded AI execution.
• Establish predictable and bounded inference performance for critical workloads.
• Isolate AI, real-time, and safety-critical applications on shared compute platforms.
• Coordinate CPU and GPU resources while maintaining timing and workload requirements.
• Support validation and certification efforts with observable, repeatable execution behavior.
  

Read the Webinar Q&A Blog >