In this paper we study controlled synchronization of networked robotic systems with dynamic uncertainties. Previous results in the literature on synchronization of nonlinear robotic systems have been primarily developed under the balanced communication graph assumption. By utilizing weighted storage functions, we demonstrate that synchronization is achievable in networked robotic systems communicating over strongly connected graphs that are not necessarily balanced. Previous results on controlled synchronization, based on adaptive and robust tracking schemes, are extended to strongly connected communication graphs. The robustness of the proposed algorithms to time delays in communication is also discussed. The control schemes are validated via simulations on a group of two-link robotic manipulators and the robustness of the proposed algorithms to noise in the system dynamics is also studied.