Efficient Convex Quadratic Optimization Solver for Embedded MPC Applications

Sammanfattning: Model predictive control (MPC) is an advanced control technique thatrequires solving an optimization problem at each sampling instant. Severalemerging applications require the use of short sampling times tocope with the fast dynamics of the underlying process. In many cases,these applications also need to be implemented on embedded hardwarewith limited resources. As a result, the use of model predictivecontrollers in these application domains remains challenging.This work deals with the implementation of an interior point algorithmfor use in embedded MPC applications. We propose a modularsoftware design that allows for high solver customization, while stillproducing compact and fast code. Our interior point method includesan efficient implementation of a novel approach to constraint softening,which has only been tested in high-level languages before. We showthat a well conceived low-level implementation of integrated constraintsoftening adds no significant overhead to the solution time, and hence,constitutes an attractive alternative in embedded MPC solvers.