We propose to developed libraries of formalized mathematics concerning linear algebra, real number computation, and algebraic topology. The main originality of this work will be to structure these libraries as a software development, relying on a basis that has already shown its power in the formal proof of the four color theorem, and to address topics that were mostly left untouched by previous research in formal proof or formal methods. The main milestones of this work will concern formally proved algorithms for solving problems in real arithmetics, solving problems in ordinary differential equations, or solving problems in algebraic topology. The success of this proposal will bring about a revolution in the way mathematicians can exploit computer-based calculations to advance their results. This research will also have an impact in all areas of software development where software is interacting with the real physical world. As a result, there should be a visible positive impact in the competitiveness of European industry. We have entered an era of mathematical proofs of extraordinary complexity that indicates a change in our understanding of mathematical reasoning. An example of such complex proofs is the four-color theorem. It involves computers in a crucial way. Can we trust such proofs? To address this problem, mathematicians and computer scientists have started research on formal proofs, and with interactive computer-based proof systems, researchers can verify algorithms whenever a complete mathematical specification is feasible. The goal of this project is to make formal proof verification available to domains that were hitherto beyond the reach of proof systems. Mathematics is already playing a crucial role in the design of sophisticated systems that are used daily, in geometrical modeling, robotics, ...This use of mathematics will increase, and correctness and reliable specification of these systems will become more and more important.