Denver Aerospace and Defense Symposium

Simulation has become an integral part of product development in the aerospace and defense industry and is reaching new levels with trends like Digital Twins. How can the power of simulation be combined with the impressive models behind the AI hype? Truth be told – in many ways! 

In this session, we will talk about four large trends for AI and Simulation: 

• AI for Reduced Order Modelling (ROM) – a way of managing the computational burden from High-Fidelity models. 
• AI-based Virtual Sensors – mimicking a physical sensor where real values cannot be measured. 
• Reinforcement Learning-based Controls – letting the model teach itself against a simulation environment. 
• AI model enhancement using Digital Twins – diversify your dataset using simulation. 

The space sector is moving away from traditional large one-of-a-kind satellite development that have historically been slow and costly.  With the formation of the Space Development Agency (SDA), new space programs are embracing faster iteration cycles, including rapid development, prototyping, and deployment of satellite constellations, to significantly reduce schedule and cost.  In this seminar, learn how to pivot to this dynamic workflow for faster mission evaluations, quick turn design iterations via simulations, and automatic tactical code generation.   

Highlights:

• Traditional development vs rapid prototyping workflows
• Mission feasibility analysis with constellation design example
• Quick-set system architecture
• Rapid design iterations via simulation
• Beyond deployment - digital twins

This presentation provides an overview of the Lockheed Martin (LM) LINUSS mission’s use of a prototype Simulink Real-Time and Speedgoat based simulator system during mission preparation and early execution. This solution realized significant cost and schedule savings during development, allowed the reuse of existing truth modeling code, and enabled the use of flight ready system images on flight EDU processor hardware.  Additionally, the use of the system during mission training and rehearsals provided highly realistic scenarios for operators.  Most importantly, the fidelity of the solution was critical in identifying, analyzing, and mitigating multiple mission-critical Flight and GNC software faults prior to experiencing issues on-orbit but after the spacecraft were mounted to launch vehicle.  As a result of these impacts the system became a key contributor to mission success.  The success of this prototype simulation system has led to wider adoption of Simulink Real-Time and Speedgoat as low-cost, high-reliability simulators at LM.

This panel explores how Digital Engineering and AI are affecting space system design, development, and operations. Executives from leading space companies and government organizations will discuss practical applications, challenges, and strategies for integrating these technologies to enhance mission success and efficiency.

In this session you will learn the capabilities of space mission modeling and analysis using Aerospace Toolbox and Blockset. It showcases the complete workflow for space system design, major features, and an overview of select in-product examples. It also provides an overview of the incremental release of major features.

Highlights

• Create a high-level view of their spacecraft with system architecture
• Derive lower-level requirements for spacecraft platform design
• Conduct mission analysis to understand and analyze mission geometry

In this hands-on workshop, you will learn how to develop AI applications using MATLAB and Simulink. Our goal is to introduce you to tools and approaches for developing predictive models on signal data. The workshop comprises three chapters, with hands- on exercises for each topic.

Highlights:

• Training and comparison of various model architectures (including LSTM and Machine Learning models) for regression
• Import of a trained model from TensorFlow
• Integration of previously trained models in Simulink and generic C/C++ code generation

The example data set we are going to use contains data for estimating the state of charge (SOC) of a battery based on temperature voltage and current