TY - JOUR JF - JoC-EE JO - jocee VL - 1 IS - 1 PY - 2022 Y1 - 2022/7/01 TI - A Novel Integrated Solution Algorithm for Explicit Model Predictive Control in Embedded Applications TT - N2 - The paper addresses complexity of explicit model predictive control (MPC) in terms of online evaluation and memory requirement. Explicit MPC defines a piecewise affine (PWA) function over different regions of system state-space. An efficient approach is presented to integrate both complexity reduction schemes via 1) a separator function to remove regions defined over control actions which attain saturated values and 2) elimination of regions which have symmetry. The proposed method reduces the conventional necessity of explicit MPC, online evaluation, and storage requirement, by removing the regions corresponding to the saturated optimal control inputs using a simpler replacement function. Moreover, the method incorporates the concept of symmetries in the context of MPC quadratic programming problem to eliminate redundant symmetric regions, leading to devising a novel solution algorithm with much less complexities for embedded applications. The presented method also simplifies the symmetry identification process because the symmetry search algorithm is performed only for regions on which control action is unsaturated rather than the whole original regions. Various simulation tests are conducted to comparatively demonstrate effectiveness of the proposed algorithm for an inexpensive implementation of large-order systems in terms of the required storage and number of floating point operations. SP - 37 EP - 47 AU - Arezoo, Jamal AU - Salahshoor, Karim AD - Professor, Department of Automation and Instrumentation, Petroleum University of Technology. KW - Explicit MPC KW - quadratic programing KW - symmetric regions KW - and saturated regions UR - http://jocee.kntu.ac.ir/article-1-33-en.html DO - 10.52547/jocee.1.1.37 ER -