This paper deals with the position synchronisation problem for a class of electromechanical systems, which can be represented by Euler-Lagrange equations, where velocities of agents are not measurable. The agents share their position errors and filter-output signals, instead of actual velocity errors, over a directed communication network, in the presence of time-varying delay. Besides, balanced communication topology is also not required. The proposed adaptive control mechanism ensures the position synchronisation among agents, as well as the tracking of each agent to given desired trajectory, under the effects of time-varying communication delay, parametric uncertainty and unknown external disturbance, simultaneously. By employing a classical Lyapunov approach and Lyapunov-Krasovskii functionals, it is concluded that the suggested control structure yields semi-global asymptotic convergence of all closed-loop errors, including synchronisation errors. The simulation results, on a six-marine vessel network system, are presented to validate the feasibility of the proposed control scheme.