Optimizing Network Topology to Reduce Aggregate Traffic in Systems of Mobile Robots

TL;DR

This paper addresses optimizing network topology in mobile robot systems to minimize total communication traffic, proposing algorithms for reconfiguration under movement constraints, with simulations comparing their effectiveness.

Contribution

It introduces algorithms for topology reconfiguration in mobile robot networks that balance optimality and computational complexity under movement constraints.

Findings

The greedy algorithm performs close to the optimal in simulations.

Topology reconfiguration maintains connectivity during movement.

Simulation results demonstrate the effectiveness of the proposed algorithms.

Abstract

Systems of networked mobile robots, such as unmanned aerial or ground vehicles, will play important roles in future military and commercial applications. The communications for such systems will typically be over wireless links and may require that the robots form an ad hoc network and communicate on a peer-to-peer basis. In this paper, we consider the problem of optimizing the network topology to minimize the total traffic in a network required to support a given set of data flows under constraints on the amount of movement possible at each mobile robot. In this paper, we consider a subclass of this problem in which the initial and final topologies are trees, and the movement restrictions are given in terms of the number of edges in the graph that must be traversed. We develop algorithms to optimize the network topology while maintaining network connectivity during the topology reconfiguration process. Our topology reconfiguration algorithm uses the concept of prefix labelling and routing to move nodes through the network while maintaining network connectivity. We develop two algorithms to determine the final network topology: an optimal, but computationally complex algorithm, and a greedy suboptimal algorithm that has much lower complexity. We present simulation results to compare the performance of these algorithm.