In this paper, we study the matrix-weighted consensus problem for Euler-Lagrange (E-L) modeled agents. Several consensus algorithms are proposed and the convergence analysis are proved based on Barbalat's lemma. More specifically, matrix-weighted consensus algorithms for E-L systems with undirected graphs, with bounded control inputs, without knowledge of the derivatives of the generalized coordinates or with uncertainties in the model of the agents will be studied in this paper. Simulation results are also provided to support the mathematical analysis.