Model Predictive Control

By Mazen Alamir, Frank Allgöwer. 2008

3,466 citations recorded by the source corpus.

Control design often seeks the best trajectory along which to move a system from its current state to a target state. Most control methods consider only the first step of the full trajectory toward the target state. Model predictive control considers the full sequence of steps required to move the system optimally from its current state to a future target. The control system then applies the first inputs to start the system along that optimal trajectory. However, rather than continuing, the system takes new inputs and recalculates a new optimal trajectory based on its updated information. The system then begins to move along the new trajectory, updates again, and adjusts its trajectory again. By repeated updating, the system can often perform very well with limited information about nonlinear dynamics and other uncertainties.

Subfields: Control and Systems Engineering, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Control and Systems Engineering

In Molecular Biology

Citations in the corpus

Composable Action-Conditioned Predictors: Flexible Off-Policy Learning for Robot Navigation (2018, methodology): “Although this MPC approach is not equivalent to planning for the entire horizon of the task, it has been shown to be an effective and robust method for robot control [32].”
Robustness via Retrying: Closed-Loop Robotic Manipulation with Self-Supervised Learning (2018, background): “Model-predictive control (MPC) [24] has proven successful in a variety of robotic tasks [25, 26, 27, 28, 29].”
An Online Learning Approach to Model Predictive Control (2019, methodology): “Various MPC algorithms have been proposed, using tools ranging from constrained optimization techniques [7, 20, 27] to sampling-based techniques [30].”
Generalized Hidden Parameter MDPs Transferable Model-based RL in a Handful of Trials (2020, methodology): “Alternatively, we use model predictive control (MPC) to find the action trajectory at:t+H that optimizes ∑t+H−1 t Eqφ(z)Ep(st,at)[p(rt+1|st,at, st+1, z)] at run-time (Camacho and Bordons 2013), using st+1 predicted from the learned model (Algorithm 1, line 10).”
Reasoning About Liquids via Closed-Loop Simulation (2017, methodology): “Using an algorithm such as model predictive control [4], the robot could plan for a short time horizon in the future using the open-loop simulation, but track the current state using the closed-loop simulation, thus preventing a fatal divergence from reality.”
Optimism-driven exploration for nonlinear systems (2015, methodology): “We control the dynamical system using model-predictive control (MPC) [14], [15], [16], [17].”
Operator Splitting for a Homogeneous Embedding of the Linear Complementarity Problem (2020, neutral): “Model predictive control with quadratic stage costs is a QCP with a particular sparsity structure [13, 59, 67].”
Trajectory-guided Control Prediction for End-to-end Autonomous Driving: A Simple yet Strong Baseline (2022, methodology): “The industry usually adopts sophisticated control algorithms such as model predictive control to achieve reliable trajectory-following performance [5, 21].”
Model Predictive Control of Power Electronic Systems: Methods, Results, and Challenges (2020, neutral): “with a finite associated cost [1], [2].”
Data-driven discovery of Koopman eigenfunctions for control (2017, neutral): “A wide range of data-driven and nonlinear control approaches exist in the literature, including model-free adaptive control [34], extremum-seeking [3], gain scheduling [55], feedback linearization [15], describing functions [68], sliding mode control [20], singular perturbation [31], geometric contr”
Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids (2018, positive): “We use Model-Predictive Control (MPC) (Camacho & Alba, 2013) to stabilize the trajectory by doing forward simulation and updating the control inputs at every time step to compensate the simulation error.”