In this paper, a guaranteed cost control for non-linear uncertain systems is developed. The relationship between the idea of guaranteed cost control and dissipative feedback is first discussed which have been independently studied in the literature on robust control. It turns out that the two approaches are equivalent, and both presume the existence of a storage function (a Lyapunov function candidate) which guarantees the stability of the closed-loop system. The idea of guaranteed cost control is then applied to solve the robust control problem for non-linear systems with both drift vector field and input vector field uncertainties. Necessary and sufficient conditions are given for the solvability of this problem in terms of certain Hamilton-Jacobi Isaac's inequalities. The robustness properties of the controllers are also discussed. Finally, a passivity-based approach to the synthesis of guaranteed cost controllers is discussed.