Agree & Join LinkedIn
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
One of the biggest assets an #ADCS engineer can have is an all-purpose simulation tool that can support distinct scenarios and compare various control law performances while still having a flexibility to tune the #simulation setup, dynamics and controller gains. This has motivated me to professionally design a MATLAB/Simulink application specifically comparing three attitude control schemes under four distinct scenarios while a chaser rendezvous with the ISS.
The attitude is modelled with quaternion kinematics and rigid-body dynamics with reaction-wheel actuation and it’s gyroscopic coupling. Direction-cosine transformations move vectors between inertial, body and reference frames, and the error quaternion is split into pointing error and twist about the docking axis. The selected controllers for the comparison are, a robust sliding-mode law, an adaptive sliding-mode law with inertia estimator (with parameter projection and anti-windup), and a quaternion PD baseline controller. It has two interchangeable simulation engines: a generated Simulink model (default), and a built-in fixed-step RK4 MATLAB reference engine, both agreeing to about 1e-11 degrees on the deterministic cases. The Simulink model is auto-generated by scripts, therefore capable of build reproducibility. A test suite of 9 automated tests passes, and a smoke test runs every scenario across all three controllers: 20 of 21 runs pass with identical verdicts on both engines.
A year ago for my research projects, I had designed these necessary MATLAB functions and Simulink block diagrams for both adaptive SMC and PD controllers for a typical co-planer along-track LEO rendezvous with exponential and sinusoidal docking axis approach profile. However, while designing the MATLAB app, formalizing my codebase to be distribution-ready and easily accessible to already existing #LLM agents turned out to be strenuous. In a separate project I have been optimizing a research & development agentic workflow as a self-sufficient orchestrator that can understand a pre-existing project and suggest/implement format checks, test suites and script/project-level documentation. With the help of that, I defined the project-level rules, guardrail hooks for #MCPs and a context-management scheme (a project guide, a versioned notes file and quality-checked external sources) so it could understand the whole project and retain relevant memory. One unique capability of this project is its mis-wire checks and repairing tools that an agent can always leverage in an agentic coding setup. Although the app is designed to integrate with any LLM agent, it still relies on certain skills I would deploy soon. As I’m ramping up my developer journey and getting comfortable with agentic development, I’m open to valuable feedback from folks already leveled with production expertise in the industry.
Link to my codebase - https://lnkd.in/em6E46kZ
#ControlSystems #GNC #Aerospace #MATLAB #Simulink #ClaudeCode #AgenticCodingkeep it up Suryansh Aryan 🥳 To view or add a comment, sign in
1,440 followers
MathWorks has announced Release 2026b (R2026b) of #MATLAB and #Simulink, introducing new products and major enhancements that enable #engineering teams to manage increasing system complexity while accelerating the path from #design and #simulation to #deployment
https://lnkd.in/eEAPZZDB To view or add a comment, sign in Read how Model-Based Design and code generation with MATLAB and Simulink were used to embed AI-powered predictive maintenance algorithms in contactors. #ModelBasedDesign #AI
https://spr.ly/6047BGtVDW To view or add a comment, sign in
2,792 followers
MathWorks has launched Release 2026b (R2026b) of MATLAB and Simulink, introducing new products and updates designed to help engineering teams manage increasingly complex systems and streamline development.
The latest release adds three new offerings: Simulink Variant Manager, Visual Inspection Toolbox, and Polyspace Test Server.
The update also brings enhancements across a range of application areas, including aerospace, embedded systems, artificial intelligence (AI), perception technologies, audio engineering and statistical analysis. Read more by using the link below 👇 https://lnkd.in/eiw3dh5e #MathWorks #newelectronics #design #MATLAB #SimuLink #engineering To view or add a comment, sign in
13,097 followers
@MathWorks announces the @MATLAB and #Simulink Release R2026b, introducing new products and enhancements for engineering teams developing increasingly complex systems. #modeling #simulation
To view or add a comment, sign in
Systems engineering is shifting from document-centric workflows to model-based, data-driven approaches that compress decision cycles from months to days, using simulation and executable models to enable faster, evidence-based design decisions.
Becky Petteys highlights how MathWorks tools like MATLAB and System Composer, together with SysML v2 and AI, improve traceability and integration by linking requirements, design, and simulation through reusable digital threads: https://spr.ly/6046BGRZ5C To view or add a comment, sign in
676 followers
Testing and Validation of Simulink® Models with Reactis®
This white paper discusses how the Reactis automatic test generation tool may be used to validate Simulink® models of embedded control software and to test for conformance of code to Simulink® models.
Reactis Tester automatically generates test cases that stress the model. The test generation often uncovers run-time errors in Simulink® models. The generated tests aim to maximize coverage with respect to a number of test coverage metrics including Modified Condition/Decision Coverage (MC/DC).
Reactis Simulator is a simulation environment for Simulink® models that enables the user to execute and debug models and to track coverage during test execution.
Reactis Validator enables an engineer to formalize model requirements as assertions and perform an automatic check for requirement violations. Validator performs these checks by thoroughly simulating the model with the goal of violating assertions. When an assertion fails, Validator returns a test that highlights the problem.
Test suites generated by Reactis serve as a testing oracle to determine if source code conforms to the behavior of a Simulink® model. The Reactis for C Plugin integrates seamlessly with Reactis to offer white-box testing for the C code portions of models (C Caller blocks, S-Functions, and Stateflow® custom code). Similarly the Reactis for EML Plugin offers white-box testing of Embedded MATLAB® within a model.
Read whitepaper → https://lnkd.in/dQ7CH2Ys To view or add a comment, sign in I’m so happy that mine and Neil MacEwen‘s application note is finally out!!! Learn how you can use Wireless Testbench to deploy algorithms to NI USRP radios and why channel emulation is a suitable application!https://lnkd.in/gtzGUR-K https://lnkd.in/gZv9iVpJ
Great to see this new blog post and application note from my colleagues Nadia Shivarova and Neil MacEwen, showcasing how MATLAB®, Simulink®, and Wireless Testbench™ can be used to develop, deploy, and validate FPGA-based wireless signal processing on NI USRP™ radios.
The article walks through a practical channel emulation example, demonstrating a seamless workflow from scenario definition and algorithm modeling to FPGA deployment and hardware-in-the-loop validation. What I find particularly compelling is how the same workflow can scale this SISO channel example to support: ✅ MIMO research ✅ Non-Terrestrial Networks (NTN) and satellite communications ✅ AI training data generation ✅ O-RAN research ✅ OAI testbeds
By maintaining continuity from simulation to real hardware, researchers and engineers can accelerate prototyping while ensuring repeatable, controlled wireless testing environments.
If you are working on wireless system design, SDR-based prototyping, FPGA implementation, NTN, or AI-driven wireless research, this is worth a read:
#WirelessEngineering #SDR #USRP #FPGA #MATLAB #Simulink #WirelessTestbench #NTN #6G #ORAN #AIforWireless #WirelessResearch To view or add a comment, sign in https://lnkd.in/gtzGUR-K https://lnkd.in/gZv9iVpJ
Great to see this new blog post and application note from my colleagues Nadia Shivarova and Neil MacEwen, showcasing how MATLAB®, Simulink®, and Wireless Testbench™ can be used to develop, deploy, and validate FPGA-based wireless signal processing on NI USRP™ radios.
The article walks through a practical channel emulation example, demonstrating a seamless workflow from scenario definition and algorithm modeling to FPGA deployment and hardware-in-the-loop validation. What I find particularly compelling is how the same workflow can scale this SISO channel example to support: ✅ MIMO research ✅ Non-Terrestrial Networks (NTN) and satellite communications ✅ AI training data generation ✅ O-RAN research ✅ OAI testbeds
By maintaining continuity from simulation to real hardware, researchers and engineers can accelerate prototyping while ensuring repeatable, controlled wireless testing environments.
If you are working on wireless system design, SDR-based prototyping, FPGA implementation, NTN, or AI-driven wireless research, this is worth a read:
#WirelessEngineering #SDR #USRP #FPGA #MATLAB #Simulink #WirelessTestbench #NTN #6G #ORAN #AIforWireless #WirelessResearch To view or add a comment, sign in 🚗🤖 Could an AI agent carry an automotive software feature from the first requirement through modeling and testing, all the way to a verified report?
Together with Levent Kent , Mohammed Hashem and the tracetronic GmbH team, we turned this question into a working proof of concept. It covers MIL (Model-in-the-Loop) and SIL (Software-in-the-Loop), the two stages where a design is tested first as a model and then as generated code, before any hardware is involved. The whole workflow runs on AWS.
Great collaboration with Levent Kent Francesco Salamida, Mohammed Hashem, and tracetronic team 🙌
If you work on embedded or automotive software, the full workflow, source code and demo video are in the original post below. I’d be glad to hear how you would use this in your own V-model process 👇
#AgenticAI #AutomotiveSoftware #Kiro #DeveloperProductivity #AWS #ModelBasedDesign #SoftwareInTheLoop #ModelInTheLoop #EmbeddedSystems Senior Solutions Architect - GenAI Automotive & Manufacturing 🚗 What if an AI agent could design, model, and verify safety-critical automotive software — end to end — on its own? That’s exactly what we built. An agentic workflow that takes requirements all the way to verified test reports across the MIL and SIL stages of embedded development, running on AWS. Here’s how the loop works 👇 1️⃣ The agent reads requirements straight from the ALM system (Polarion) via an MCP server 2️⃣ It builds a Simulink plant model from those requirements via Simulink Agentic Toolkit 3️⃣ It runs the simulation and checks results against the requirements (Model-in-the-Loop) 4️⃣ It exports the model as an FMI 2.0 Co-Simulation FMU and ships it to the cloud 5️⃣ It cross-compiles the FMU’s C sources for Linux (Software-in-the-Loop) 6️⃣–7️⃣ It drives ecu.test working with the REST API and uses tracetronic GenAI to auto-generate the test scripts 8️⃣ It executes the full test suite against the FMU and produces test reports 9️⃣ And Kiro iterates — adjusting the model until the feature is fully designed, implemented, and verified The part I love: it stitches together some of the most widely used tools in automotive engineering — Siemens Polarion for requirements and traceability, MATLAB/Simulink for modeling, and ecu.test for verification — inside a single autonomous agent loop, powered by Kiro. This is where agentic coding meets real engineering rigor. And the same pattern opens up unblocks much more: imagine each engineering tool virtualized in its own instances/containers, with the agent exploring model alternatives in parallel and converging on the best one automatically.
See the source code here: https://github.com/LeventKent/automotive-sdlc-agent-for-mil-sil-hil Demo: https://github.com/LeventKent/automotive-sdlc-agent-for-mil-sil-hil/blob/main/MiL-SiL-Demo.mp4
#AgenticAI #AutomotiveSoftware #SDLC #ModelBasedDesign #AWS #Kiro #ecu.test #Simulink #Polarion #SoftwareInTheLoop #ModelInTheLoop #EmbeddedSystems
Special thanks goes to: AWS side: Mohammed Hashem, Francesco Salamida and Oktay Esgul and tracetronic GmbH team: Sebastian Dietze, Maximilian Krey, Patrick Stiller
FYI: Stefano Marzani, Mahesh Geeniga, Yuri Lukin, Danny Smith To view or add a comment, sign in This Workflow is an example for the excellent partnership between tracetronic and Amazon Web Services (AWS). Building high-quality models is a major step in automotive software testing. In the future, we want to extend this workflow to higher SiL levels or full virtual vehicles.
See the source code here: https://github.com/LeventKent/automotive-sdlc-agent-for-mil-sil-hil Demo: https://github.com/LeventKent/automotive-sdlc-agent-for-mil-sil-hil/blob/main/MiL-SiL-Demo.mp4
#AgenticAI #AutomotiveSoftware #SDLC #ModelBasedDesign #AWS #Kiro #ecu.test #Simulink #Polarion #SoftwareInTheLoop #ModelInTheLoop #EmbeddedSystems
Special thanks goes to: AWS side: Mohammed Hashem, Francesco Salamida and Oktay Esgul and tracetronic GmbH team: Sebastian Dietze, Maximilian Krey, Patrick Stiller
FYI: Stefano Marzani, Mahesh Geeniga, Yuri Lukin, Danny Smith To view or add a comment, sign in
981 followers
Create your free account or sign in to continue your search
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.