Model Predictive Control for Automated Driving
Model predictive control (MPC) is a popular technology for implementing as adaptive cruise control, lane keeping assist, path following control, automated parking and other automated driving algorithms.
In this session, you will learn how you can quickly get started with developing these algorithms by using prebuilt automated driving blocks and reference examples. You can customize the references examples to design and deploy MPC controllers for your own automated driving application or design MPC algorithms from scratch. You can develop and simulate scenes, sensors and virtual vehicle models, create different driving scenarios, and test MPC algorithms in simulated 3D environments. You will hear about experiences of companies who developed and implemented various automated driving algorithms using MPC. You will also learn how to design, simulate, and implement linear, adaptive and nonlinear MPC controllers, and integrate them with perception and path planning algorithms.
This event is part of a series of related topics. To view the full list of events in this series click here.
- MathWorks solutions for developing ADAS applications using MPC
- MPC basics: What it is, why use it, how it works
- Demonstrations of linear, adaptive and nonlinear MPC design for
lane following and parking valet applications
- Automated driving and Model Predictive Control resources.
Who Should Attend
Educators, Researchers, Students, Racing Teams
About the Presenter
Don Bradfield is a senior application engineer at MathWorks on the automated driving product focused team. He graduated from Penn State with a bachelor’s in mechanical engineering and spent 6.5 years at Ford Motor Company as a systems engineer focusing on autonomous vehicle motion performance & control and self-driving ride quality. His areas of expertise are motion control, vehicle-level simulation, data post-processing, and in-vehicle evaluation. Don has several patents in the automotive space.