Learning Based Model Predictive Control
Research Area : Machine Learning, Model Predictive Control, Embedded Control.
Saket Adhau
Guide : Dr. D. N. Sonawane
Year : 2017-2019
Mail : [email protected]
My Profile
Research Area : Machine Learning, Model Predictive Control, Embedded Control.
Saket Adhau
Guide : Dr. D. N. Sonawane
Year : 2017-2019
Mail : [email protected]
My Profile
Abstract -
MPC is applicable whenever a process has a dead time dominant dynamics, whenever there are interacting process loops in conventional controls where you want to apply feed forward. Model predictive control technology is a good vehicle for optimization and with optimization you can get some economic benefits in addition to just improve control performance. Those economic benefits can be achieving objectives like minimizing energy consumption, maximizing production, minimization of undesirable by products.
The Learning Model Predictive Control (LMPC) framework combines model-based control strategy and machine learning technique to provide a simple and systematic strategy to improve the control design using the plant data. LBMPC deals with the uncertainties of the learning model, as well as the measurement noises, and also satisfies the required conditions to ensure the robustness of the controlling commands. The beneficial effect of introducing learning to the MPC control is precisely to enrich the level of representation of the model in order to obtain the performances with the great precision.
Until now, MPC and EMPC have been successfully implemented on ARM Cortex platform, the next task being the successful implementation of LBMPC.
MPC is applicable whenever a process has a dead time dominant dynamics, whenever there are interacting process loops in conventional controls where you want to apply feed forward. Model predictive control technology is a good vehicle for optimization and with optimization you can get some economic benefits in addition to just improve control performance. Those economic benefits can be achieving objectives like minimizing energy consumption, maximizing production, minimization of undesirable by products.
The Learning Model Predictive Control (LMPC) framework combines model-based control strategy and machine learning technique to provide a simple and systematic strategy to improve the control design using the plant data. LBMPC deals with the uncertainties of the learning model, as well as the measurement noises, and also satisfies the required conditions to ensure the robustness of the controlling commands. The beneficial effect of introducing learning to the MPC control is precisely to enrich the level of representation of the model in order to obtain the performances with the great precision.
Until now, MPC and EMPC have been successfully implemented on ARM Cortex platform, the next task being the successful implementation of LBMPC.
