In this paper, a new focal quality indicator, the 2D-TKEO indicator, for the translational motion compensation is proposed. This indicator is based on the 2D Teager-Kaiser Energy Operator which represents the 'energy' of an image. In Inverse Synthetic Aperture Radar (ISAR) imaging of a non-cooperative target, the relative motion is unknown and it is therefore necessary to implement an autofocus approach to obtain high resolution imaging. This step is based on an optimization problem via the minimization of a cost function, thus making it possible to estimate the translational kinematic (speed, acceleration, jerk.. .) of the moving target. This paper describes the use of a new quadratic focal quality indicator to estimate the quality of an ISAR image. To validate our approach is presented on simulation results of a target constituted by ideal point scatterers models. The effectiveness of 2D-TKEO indicator is demonstrated on both noise-free and noisy simulations and results compared to current methods.