SwarmMesh: A Distributed Data Structure for Cooperative Multi-Robot Applications

Abstract

We present an approach to the distributed storage of data across a swarm of mobile robots that forms a shared global memory. We assume that external storage infrastructure is absent, and that each robot is capable of devoting a quota of memory and bandwidth to distributed storage. Our approach is motivated by the insight that in many applications data is collected at the periphery of a swarm topology, but the periphery also happens to be the most dangerous location for storing data, especially in exploration missions. Our approach is designed to promote data storage in the locations in the swarm that best suit a specific feature of interest in the data, while accounting for the constantly changing topology due to individual motion. We analyze two possible features of interest: the data type and the data item position in the environment. We assess the performance of our approach in a large set of simulated experiments. The evaluation shows that our approach is capable of storing quantities of data that exceed the memory of individual robots, while maintaining near-perfect data retention in high-load conditions.

DORA: Distributed Online Risk-Aware Explorer

“…Also, more experiments could be conducted by testing DORA Explorer on a larger team of physical robots exploring larger outdoor environments. Further applications of DORA could include using the generated risk belief map to determine robots' fitness to store data in distributed storage systems like SwarmMesh [11], with robots assigned to tasks in dangerous regions being discouraged from storing sensitive information. Additionally, in this work we considered that the risk associated with the environment can be sensed by the robots.…”

“…This solution differs from distributed hash tables, which require a complete view of the system at every point in time. Other distributed data storage approaches such as SwarmMesh [11] store data in different locations based on a fitness function instead of replicating them on all robots. This allows the storage of more data with less communication, but robots are less likely to have access to the latest values.…”

RASS: Risk-Aware Swarm Storage

“…A memory-efficient storage mechanism designed for multi-robot systems was proposed in SwarmMesh [8]. It uses the available memory of an agent and its connectivity with its peers to generate unique identifiers, which are the basis of its partitioning scheme and which represent an agent's fitness to store data.…”

“…We implement message forwarding through a gossip algorithm, which allows local broadcasting between robots. 2) Potential-Based Percolation: Our risk-aware storage system draws inspiration from SwarmMesh [8], in that each node periodically assigns itself a potential φ i based on its fitness to store data, given by:…”

“…Parameters α and β are respectively the routing weights and risk weights, which allow adapting the policy based on the relative importance of the routing time and the environmental, risk with respect to each other. Similarly to [8], a node which becomes unfit to store data will evict such data by moving it into its routing queue. The condition for "unfitness" is simply:…”