Abstract: In order to beat the opponent in a competitive networked robot game, the robot need hit the target quickly and accurately while avoiding enemy's attack. In this paper, a novel approach combining trajectory planning and image-based visual servo control is proposed to quickly and effectively meet the requirements. Some main constraints such as attack avoiding, visibility, kinematic singularities, and joint velocity limits are taken into account. A trajectory is designed in the image space, which can be used in the image-based visual control directly. Simulations and experiments using an eye-in-hand robotic system on a 6-DOF (degree of freedom) robot manipulator are carried out to confirm that our approach can help the robot to hit the target and avoid attacks.