This paper proposes a novel cost function formulation for minimization of the energy consumption of industrial robots by trajectory optimization. Besides the dynamics of the robot including friction losses, the model especially takes into account the losses of servo drives and inverters. Furthermore, the ability of energy exchange between the robot axes via the coupled DC-bus is included, since the servo drives support generator mode during deceleration. The utilized energy-based robot model is applicable to different manipulator types. For the energy-efficient motion planning, point-to-point trajectories are defined by B-spline functions. The given nonlinear optimization problem is solved using gradient-based methods, considering kinematic and dynamic constraints. Several simulation results are presented, demonstrating the intense effect of energy exchange in the robot controller’s power electronics. Furthermore, a comparative study is given showing that the proposed method is able to outperform existing torque-based approaches.