Distributed path planning for connectivity under uncertainty by ant colony optimization

Movement and allocation of network resources for a system of communicating agents are usually optimized independently. Path planning under kinematic restrictions and obstacle avoidance provides a set of paths for the agents, and given the paths, it is then the job of network design algorithms to allocate communication resources to ensure a satisfactory rate of information exchange. In this paper, we consider the multiobjective problem of path planning for the sometimes conflicting goals of fast travel time and good network performance. In previous work we considered this problem under the assumption of full knowledge of network topologies and unlimited computational resources. In this paper, nothing is known a priori about topology, information is exchanged between nodes within a connected component of the network, and sources of environment-dependent communication failure can only be approximately estimated through learning. All the planning must be done online in a distributed fashion. We apply ant colony optimization to this problem of planning under uncertain information, and show that significant benefit in network performance can be achieved even under the difficult conditions of the scenario. Furthermore, we show the ability of nodes to quickly learn the communication patterns of the arena, and use this information for improved path planning.