SwarmDAG: A Partition Tolerant Distributed Ledger Protocol for Swarm Robotics

. Blockchain technology has the potential to disrupt applications beyond cryptocur-rencies. This work applies the concepts of blockchain technology to swarm robotics applications. Swarm robots typically operate in a distributed fashion, wherein the collaboration and coordination between the robots are essential to accomplishing the application goals. However, robot swarms may experience network partitions either due to navigational and communication challenges or in order to perform certain tasks efﬁciently. We propose a novel protocol, SwarmDAG, that enables the maintenance of a distributed ledger based on the concept of extended virtual synchrony while managing and tolerating network partitions.

l e d g e r j o u r n a l .o r g ISSN 2379-5980 (online) associated article DOI 10.5915/LEDGER.2019.174ii information when separate partitions merge through a novel ledger structure based on a directed acyclic graph.
The authors describe the proposed protocol and list the assumptions.There are a few figures that illustrate the concept; more example with robots in motion would help strengthen the idea.
The paper is well written and structured.Although it is clearly at a premature stage (no empirical results nor theoretical proofs), it is relevant to this community and provides an interesting basis for discussion at an intimate conference such as BROS.

Reviewer B:
This paper presents SwarmDAG, a distributed ledger that integrates mechanisms to cope with partition tolerance and ensure eventual consistency.
The paper is central to the topic of the symposium.The paper is well-written and easy to follow.
I enjoyed reading this paper as I share with the authors an ongoing interest in partitiontolerant distributed data structures.I think that this work is a good discussion point for the symposium.
The main limitation of this paper is the extremely preliminary state at which it actually is.The authors do not seem to have implemented their approach, and the paper seems to describe a design proposal more than an actual system.

1B. Authors' Response
Reviewer A: The paper proposes a method that allows to maintain blockchain technology on a swarm of robots that may not be sufficiently connected at all times.The key contribution is the proposal of a protocol that the authors call 'swarmDAG' that facilitates the operation of a distributed ledger.The authors adopt the Extended Virtual Synchrony method.This allows network partitions to continue updating their progress.They address the problem of conflicting information when separate partitions merge through a novel ledger structure based on a directed acyclic graph.
The authors describe the proposed protocol and list the assumptions.There are a few figures that illustrate the concept; more example with robots in motion would help strengthen the idea.
The paper is well written and structured.Although it is clearly at a premature stage (no empirical results nor theoretical proofs), it is relevant to this community and provides an interesting basis for discussion at an intimate conference such as BROS.
 The authors thank Reviewer A for the feedback and agree that this work is relevant to the community and will spark interesting discussions at BROS.This paper presents iii the conceptual design of SwarmDAG, a distributed ledger protocol which aims to address the problem of network partitions in robotic swarms.The work presented is ongoing and the implementation of SwarmDAG is currently in progress which will be open sourced when complete.For portability and ease of reproducing the implementation, we are implementing our system components using Docker containers and focusing on compatibility for Raspberry Pis (ARM), which is commonly used in robots in the research community.The underlying blockchain toolkit in our current implementation is Tendermint, and to test for design correctness in the face of partitions, we are utilizing Blockade (https://github.com/worstcase/blockade) to emulate network partitions.To implement the Membership Management Service (MMS) as detailed in the paper, we are using the libp2p network stack (https://libp2p.io/) to facilitate peer-to-peer communication via gossip messages.In future work, we will present empirical results of the SwarmDAG implementation. In addition to the implementation, we will present a formal proof of SwarmDAG's partition tolerance in future work.Because the implementation is currently in progress, we do not have the current capability of illustrating SwarmDAG using moving robots.
We intend to run SwarmDAG on our in-house Intelligent Robotic Internet of Things TeStbed (IRIShttps://github.com/ANRGUSC/iris-testbed/) built by the authors in previous work to illustrate the effectiveness of SwarmDAG in a real swarm of robots.

Reviewer B:
This paper presents SwarmDAG, a distributed ledger that integrates mechanisms to cope with partition tolerance and ensure eventual consistency.
The paper is central to the topic of the symposium.The paper is well-written and easy to follow.
I enjoyed reading this paper as I share with the authors an ongoing interest in partitiontolerant distributed data structures.I think that this work is a good discussion point for the symposium.
The main limitation of this paper is the extremely preliminary state at which it actually is.The authors do not seem to have implemented their approach, and the paper seems to describe a design proposal more than an actual system.
 The authors appreciate the positive feedback from Reviewer B, and agree the work is at an extremely preliminary state.Due to the page limit of this venue, we were unable to fit the additional progress of the ongoing work presented in this paper.We currently have an implementation of SwarmDAG in progress, which we will open source upon completion.We also intend to present the progress of this implementation at BROS.To facilitate ease of reproducibility and portability, we are implementing SwarmDAG using Docker containers and focusing on compatibility with Raspberry Pis, a computing platform widely used by robots in l e d g e r j o u r n a l .o r g ISSN 2379-5980 (online) associated article DOI 10.5915/LEDGER.2019.174