Industrial robots are extensively used nowadays in material-handling applications. The traditional environment, in which these robots are put to work, however, requires that the motion of the objects to be handled be knowna priori, namely, be predictable. Using such advance knowledge of the motion of the moving object helps in optimizing its interception time. One can argue that a fast and reliable trajectory planner that assumes no prior information regarding the motion of the object would allow a higher level of flexibility in the design of manufacturing environments. The purpose of this workis, thus, to introduce a novel on-line trajectory-planning algorithm that results in fast robotic interception of moving objects.

There have been two approaches for addressing the on-line robotic interception problem of fast maneuvering objects. In visual-servoing techniques, a conventional tracking algorithm is in command throughout the whole interception period. These algorithms are notvery efficient, since they constantly attempt to match the velocity of the robot with that of the object even when the end-effector is far from the object. As an alternative, navigation-based techniquesresult in shorter interception times. However,in these methods, switching must occur to a conventional tracking algorithm as the end-effector approaches the object dueto the inability of the navigational methods to ensure a smooth grasp of the object. The novelty of the proposed methodis in its single-phase navigational approach, while using the robot dynamics in the selection of velocity commandsto ensure a fast and smooth interception. As a consequenceofthe single-phase nature of the method, the problems associated with the determination of the optimal switching point in navigation based techniquesare eliminated.

In the proposed methodology,first a parallel-navigation rule, originally introduced in the missile-guidance literature, is applied to generate a set of instantaneous task-space velocity commands, which, when executed, would keep the end-effector on a collision course with the object. Subsequently, a rendezvous-guidance method is utilized to reduce the original command set to one with velocity-matching capability. Finally, the fastest velocity commandin the reduced set is chosen such that the dynamic limitations of the actuators of the robot are not violated.

The proposed method has been simulated in 2D and 3D workspaces. Results were obtained for a variety of object motions and different robot parameters and initial configurations. Interception times achieved with the proposed algorithm are shorter than those obtained through conventional visual-servoing methods.