Preconditioning of Iterative Solvers for embedded real-time model predictive control

Model predictive control (MPC) for linear dynamical systems requires solving an optimal control structured quadratic program at each sampling instant. The need for embedded implementation makes developing efficient algorithms challenging, but also crucial for real-time high speed control, e.g., in automotive and robotics applications.

I am delighted to see my former colleagues Rien Quirynen and Stefano Di Cairano building up on our past research and publishing their latest work "PRESAS: Block-Structured Preconditioning of Iterative Solvers within a Primal Active-Set Method for fast MPC."

From the abstract: "Based on a standalone C code implementation, we illustrate the computational performance of PRESAS against current state of the art QP solvers for multiple linear and nonlinear MPC case studies. We also show that the solver is real-time feasible on a dSPACE MicroAutoBox-II rapid prototyping unit for vehicle control applications, and numerical reliability is illustrated based on experimental results from a test bench of small-scale autonomous vehicles."