The efficiency of distributed sensor networks depends on an optimal trade-off between the usage of resources and data quality. The work in this paper addresses the problem of optimizing this trade-off in a self-configured distributed robotic sensor network, with respect to a user-defined objective function. We investigate a quadtree network topology and implement a fully distributed threshold-based field estimation algorithm. Simulations with field data as well as real robot experiments are performed, validating our distributed control strategy and evaluating the threshold-based formula for real world scenarios. We propose a theoretical analysis that predicts the system's behavior in real world case studies. The experiments and this prediction show very good correspondence, enabling the accurate employment of the objective function, optimizing the trade-off based on user needs.

Simulation is frequently used in the study of multi-agent systems. Unfortunately, in many cases, it is not necessarily clear how faithfully the details of the simulated model represent the behavior of the physical system. Often, the effects of the environment in which the system is to be placed are even neglected entirely. Taking into account theentiresystem (including interactions with the target environment), establishing a clear hierarchy amongmultiplelevels of modeling not only enhances the fidelity of the individual models, but also emphasizes the tradeoffs inherent in each. Understanding and leveraging the full spectrum of models allows the use of fast, high-level models for exploration in the parameter space, the results of which can be verified on more precise low-level models. Here, we demonstrate the generation of a family of models for a robotic wireless sensor network engaged in an acoustic detection task. Quantitative correspondence is shown between modeling levels and with the physical system.

Vision-based tracking is used in nearly all robotic laboratories for monitoring and extracting of agent positions, orientations, and trajectories. However, there is currently no accepted standard software solution available, so many research groups resort to developing and using their own custom software. In this paper, we present version 4 of SwisTrack, an open source project for simultaneous tracking of multiple agents. While its broad range of pre-implemented algorithmic components allows it to be used in a variety of experimental applications, its novelty stands in its highly modular architecture. Advanced users can therefore also implement additional customized modules which extend the functionality of the existing components within the provided interface. This paper introduces SwisTrack and shows experiments with both marked and marker-less agents.

In the present study, we are interested in verifying how the progressive addition of constraints on communication and localization impact the performance of a swarm of small robots in shape formation tasks. Identified to be of importance in a swarm-user interaction context, the time required to construct a given spatial configuration is considered as a performance metric. The experimental work reported in this paper starts from global and synchronized localization information, shown to be successful both on a real hardware system and in simulation. In a second step, communication is constrained to a local scale, thus obliging a single designated robot to disseminate the global localization information to the other agents. The reliability of the radio communication channel and its impact upon the performance of the system are considered.

Given the rigid energetic constraints under which a sensor network must operate, efficient means of power management are vital to the success of any sensor network deployment, particularly those in rapidly changing environments. Threshold-based algorithms provide a possible in-network method for adaptive distributed control of energy consumption.