This paper presents a novel method for the time-optimal interception of moving objects in the presence of obstacles. The proposed method not only provides time-optimal positional interception, but, also simultaneously matches the velocity at which the object is moving in a dynamic environment with static and/or moving obstacles: A rendezvous-guidance technique is utilized in conjunction with a modified cell-decomposition method. Simulation and experimental results clearly demonstrate the efficiency of the proposed interception method.

This paper presents a novel method for the time-optimal rendezvous of an autonomous vehicle with moving targets in the presence of dynamic obstacles. This interception problem is addressed by utilizing the velocity obstacle approach of Fiorni and Shiller augmented with a rendezvous-guidance algorithm. However, the velocity obstacle approach had to be modified to ensure that the real-time planned path deviates minimally from the one generated by the rendezvous-guidance algorithm. Simulation and experimental results, some of which are presented in this paper, clearly demonstrate the efficiency of the proposed interception method.

A measure is proposed that is based on two key concepts: (1) the congestion caused by a virtual circuit (VC) on the others; and (2) the penalty inflicted by the network on a VC. These two concepts are then combined to define a fairness criterion that uses the principle that a virtual circuit should not be able to congest the network more than it is being penalized by the same network. A suboptimal algorithm to select window sizes according to this fairness criterion was devised and applied to several networks; results are presented for two of these networks. The examples show that the window set produced by the algorithm does not cause any significant degradation in the net overall performance and is superior, in terms of fairness, to other window assignments methods discussed in the literature.>