- 6
- 52 215
Alberto Bemporad
เข้าร่วมเมื่อ 2 ก.พ. 2013
Model Predictive Control for Automotive Production
Model Predictive Control (MPC) is a highly effective advanced control technology for optimal coordination of multiple actuators under constraints on input and output variables. In this talk, I describe the fundamental principles of MPC and offer insights into recent advancements in learning-based methods for both MPC design and calibration. I also highlight MPC's key advantages over conventional methods, such as classical PID control, in terms of superior multi-actuator coordination, reduced setup and calibration effort, and ease of maintenance and design portability, providing compelling evidence of MPC's feasibility for mass production in the automotive industry.
Slides: cse.lab.imtlucca.it/~bemporad/talks/mpc_9_11_2023.pdf
Slides: cse.lab.imtlucca.it/~bemporad/talks/mpc_9_11_2023.pdf
มุมมอง: 942
วีดีโอ
MPC from Basics to Learning-based Design (2/2)
มุมมอง 7K2 ปีที่แล้ว
Lecture at the First ELO-X Seasonal School and Workshop (March 22, 2022). Contents of this video: - Learning-based nonlinear MPC (feedforward and recurrent neural networks) - Learning-based hybrid MPC (piecewise affine models) - Active preference learning for MPC calibration Slides: cse.lab.imtlucca.it/~bemporad/talks/elox2022-bemporad.pdf
MPC from Basics to Learning-based Design (1/2)
มุมมอง 28K2 ปีที่แล้ว
Lecture at the First ELO-X Seasonal School and Workshop (March 22, 2022). Contents of this video: - Model predictive control (MPC): basic concepts - Linear MPC and extensions to nonlinear MPC - Embedded quadratic optimization Slides: cse.lab.imtlucca.it/~bemporad/talks/elox2022-bemporad.pdf
Machine Learning Methods for Model Predictive Control
มุมมอง 8K3 ปีที่แล้ว
Semi-plenary lecture by Alberto Bemporad at the European Control Conference 2021, July 2, 2021. Slides: cse.lab.imtlucca.it/~bemporad/talks/ecc2021.pdf ABSTRACT: Machine learning is a set of techniques to extract mathematical models from data that has recently become extremely popular and very successful in many fields, including control. In my talk, I will present several approaches in which m...
Model Predictive Control: A Rising Technology in the Automotive Industry
มุมมอง 4.9K4 ปีที่แล้ว
Plenary lecture by Alberto Bemporad at the 4th IEEE Conference on Control Technology and Applications (CCTA 2020), Montreal, Canada, August 25, 2020. Slides: cse.lab.imtlucca.it/~bemporad/talks/ccta2020/bemporad-ccta2020.pdf ABSTRACT: Model Predictive Control (MPC) is widely recognized as a very effective advanced control technique for multivariable systems subject to constraints on input and o...
Machine Learning: A New ICE (Identification, Control, Estimation) Age ?
มุมมอง 2.5K4 ปีที่แล้ว
Talk by Alberto Bemporad at the IFAC 2020 Workshop on "Machine Learning meets Model-based Control", July 11, 2020. Slides: cse.lab.imtlucca.it/~bemporad/talks/ifac2020/bemporad-ifac2020.pdf ABSTRACT: Control theory always evolved by taking full advantage of developments in other disciplines, for example frequency-domain methods have leveraged on complex analysis, state-space approaches on linea...
Comment first
Thank you, it's very clear and easy to understand!
Thank you for making this video Sir. May I request you to make a video on how to apply MPC in DC-DC converters please, taking buck converter as an example.
The chess analogy is good but not accurate, while chess has 16 different controller(pieces) and thousand paths but a vehicle has 2 controller and one path.
Muchas gracias por compartir.
Thank you
Thnx professor. Would you explain difference between Explicit and Implicit MPC clearly.
Thank you for sharing this great lecture! I find a bit confusing that you use the NRMSE in the sense of "fit", i.e. 100% when a model perfectly fits data. From my experience and according to the Wikipedia (NRMSD), the NRMSE is the RMSE normalized by the mean value or range, i.e. NRMSE=0% when a model perfectly fits data.
Any help on how can I implement MPC on controlling the output force of a pneumatic frictionless cylinder. I want a constant force output regardless of compression
hi, where is the process of optimizing iteration? Seems you trained the prediction model by using RNN, but in the following, I didn't see the explicit control(search for optimal u(t)) process. Could you be brief to the point? thank you.
Amend for my question, in slide TRAINING RNNS BY SEQUENTIAL LEAST-SQUARES you bring controller in, is that controller merged in RNN training process?
Found this very useful. Thank you for sharing the lecture.
Doing my master thesis in a local motion planner for automotive in urban environment, using MPC for trajectory planning & obstacle avoidance and using another MPC for trajectory tracking & vehicle stability. This video was very helpful, thank you!
Ricordati che hai rovinato insieme ad altri la vita del prof. Andrea Giotti
Many thanks
Thanks a lot
Very Informative, Thank you !
Thanks for your effort!
Pᵣₒmₒˢᵐ
Really nice tutorial. I like the second part in particular. This would be the best video I can find to give a systematic overview of data-driven and learning-based MPC. Thanks for your effort!
Thank u! Very interesting