The Convex Optimization for Real-time Embedded Systems (CORES) project aims to develop highly efficient, real-time convex optimization algorithms and toolsets for solving important engineering problems on hierarchical and heterogeneous embedded system architectures. Though recent advances in optimization solvers have enabled the solution of optimization problems on low-cost embedded systems, the size of the problems that can be solved in real-time is still limited.
The CORES project, a collaboration between ICL and Michigan Technological University, works to address this limitation. The ICL team’s main responsibility is the design and development of higher-performance, structure-aware linear solvers that would enable us to solve, in real-time, the convex optimization problems that have significantly higher performance—and are orders of magnitude greater in size—compared to current state-of-the-art solvers.
In Collaboration With
- Michigan Technological University