A Survey of Parallel Implementations for Model Predictive Control

Abstract

Model Predictive Control (MPC) has its reputation since it can handle multiple inputs and outputs with consideration to constraints. However, this comes at the cost of high computational complexity, which limits MPC to slow dynamic systems. This paper provides an overview of the available methods to accelerate the MPC process. Various parallel computing approaches using different technologies were proposed to speed up the execution of MPC, some of these approaches are focused on building dedicated hardware for MPC using field programmable arrays (FPGA), and others are focused on parallelizing MPC computation using multi-core processors (CPUs) and many-core processors (GPUs). The focus of this survey is to review the available methods for accelerating MPC process. A brief introduction to the theory of MPC is provided first followed by a description of each approach. A comparison between the different methods is presented in terms of complexity and performance followed by a valid application for each approach. Finally, this paper discusses the challenges and requirements of MPC for future applications.

Multi-Objective Optimal Model Predictive Control for Three-Level ANPC Grid-Connected Inverter

“…where k is the serial number of sampling in the system. Without considering the effect of stray resistance, the predicted value of i α,β (k + 1) can be deduced by substituting equation 3into equation (1) and computing the last extrapolate:…”

“…The world is witnessing an energy reform driven by environmental and resource constraints. Many concrete efforts have been paid to implement the reform, including building smart grids, setting up an energy Internet, and developing low-carbon renewable energies [1]- [8]. Photovoltaic (PV) power and wind power are two popular kinds of such energies [9].…”

Real-Time Implementation of Randomized Model Predictive Control for Autonomous Driving

“…The first category is the development of the sampling based solver for MPC [5], [22], [23]. The second one is the study focusing on the implementation of Sampling based MPC [24]- [28]. In the first category, [22] developed a method that searches the control space by seeding a tree from the previous best sample and expanding it in a manner similar to rapidly exploring random tree (RRT).…”

“…The optimization algorithms used were not sampling based in nature and the application domain was much different from the control of an autonomous vehicle. Finally, in the survey paper [28], a summary of various parallel implementations of MPC was described. This paper presents a detailed discussion of the present challenges that MPC faces and how parallel computation helps to solve those challenges.…”

A Loop Pairing Method for Non-Linear Multivariable Control Systems Under a Multi-Objective Optimization Approach

“…Another disadvantage is its high computational cost, as MPC generally requires to solve an optimization problem in real time. This aspect is becoming less and less problematic due to the advance of computer systems [5], [6].…”