Elsevier

Information Sciences

Volume 575, October 2021, Pages 528-541
Information Sciences

An improved DPoS consensus mechanism in blockchain based on PLTS for the smart autonomous multi-robot system

https://doi.org/10.1016/j.ins.2021.06.046Get rights and content

Highlights

Abstract

Due to the development of robot technology, smart autonomous multi-robot systems face different security problems such as data loss and vulnerabilities. However, the robot information stored in blockchain can be more transparent and effective at ensuring the security of the robot system due to the decentralization, tamper-proof and anonymity of blockchain. In the architectural composition of blockchain, Delegated Proof of Stake (DPoS) consensus mechanism is playing a critical role with more decentralization, lower energy consumption and faster confirmation speed. Similar to the board voting, the holders cast a certain number of delegates to perform verification and block generating on their behalf in DPoS. In order to improve the efficiency and flexibility of DPoS consensus mechanism, we propose an improved DPoS consensus mechanism based on the Probabilistic Linguistic Term Set (PLTS) for the smart autonomous multi-robot system. By adding voting options for nodes, the Voting Algorithm with Probabilistic Linguistic Information (VAPLI) calculates the score and deviation degree of each node after tabulating the voting results. The selection of a delegate is based on the comparison of the score and deviation degree. Finally, we explore the model implementation of the improved DPoS consensus mechanism, and verify its feasibility and effectiveness using examples.

Introduction

Contemporarily, smart robots have begun to enter new working environments, such as transportation [1], health care [2], [48], cloud computing [49], and other service industrials. According to the current development direction and security situation of smart autonomous multi-robot systems, robot applications also face new security challenges [3]. For instance, information about robot system agents may be exposed on the public Internet [4], [5], allowing attackers to maliciously break into the system and create vulnerabilities that can cause the multi-robot system to deviate from its mission [6], [7], [47]. In addition, blockchain has become a promising technology for data security management due to its immutability and tamper-resistance [8], [9]. By storing the critical information of an autonomous robot in blockchain, the data will not be spread and abused, preventing possible data risks, and improving the security and resistance of the multi-robot system to attack.

Blockchain is essentially a distributed system that is broadcast via a Peer-to-Peer (P2P) network [10], [11], [12], and creates a distributed ledger containing traceable and append-only transaction records to keep data records consistent across nodes [13], [14]. The core of blockchain is to establish a relationship of trust through a distributed network, time-immutable cryptographic ledger, distributed consensus mechanism, and smart contracts, which finally realizes the evolution from information interconnection to value interconnection in blockchain system [15], such as Hyperledger Fabric [16] and Ethereum [17]. Blockchain is the underlying technology of digital currency that securely records transaction information for all currencies in a decentralized distributed ledger [18], [19]. Combining the advantages of current symmetrical encryption, asymmetric encryption and hash algorithms, blockchain uses digital signature to ensure the security of transactions [20] and uses smart contracts composed of automated scripted code to manipulate data. Furthermore, in blockchain network, the blocks are validated, synchronized and created across nodes via a distributed, and decentralized consensus mechanism [21], [22], [23].

The consensus mechanism is the key of the entire blockchain system, since its efficiency directly determines the blockchain’s performance as data consistency [24]. Cynthia Dwork and Moni Naor first proposed the Proof of Work (PoW) mechanism [25] which depends on the computational capacity. With the continuous improvement of blockchain, other mechanisms are also emerging, such as Proof of Stake (PoS) mechanism [26], Delegated Proof of Stake (DPoS) mechanism [27], and Practical Byzantine Fault Tolerance (PBFT) mechanism [28].

The consensus mechanism describes the process by which the nodes compete for accounting rights that generate new blocks to obtain rewards [29], [30]. In the process of competing for accounting rights, PoW greatly consumes computer resources, and block generation is inefficient. Although PoS compensates for some of the deficiencies of PoW by using stakes in the competition, it is still too resource-intensive and overreliant on computing power. Therefore, DPoS introduces a voting mechanism based on PoS that reduces the time costs of block generation to some extent [27], [31]. However, the normal operation of blockchain depends on delegates in DPoS that are generated by the vote of all network nodes and are also the actual controllers used to generate blocks. The architecture of blockchain and the process of selecting delegates among the voting nodes in DPoS are depicted in Fig. 1. The role of delegates generating blocks for the smart autonomous multi-robot system is shown in Fig. 2.

Therefore, the voting selection of the delegates is the key in DPoS [32], [33], [34]. All nodes can only vote in favor without other complicated voting choices in the process of selecting delegates in the traditional DPoS consensus mechanism [31], [35], [36]. Liu et al. in [37] proposed an improved DPoS consensus mechanism based on the vague set, in which nodes can vote “yes”, “no” and “abstention”. However, there are still some issues to be addressed as follows:

  • There is too much information on smart autonomous robots in a multi-robot system, which leads to a low consensus reaching process for the nodes in blockchain, contributing to the low efficiency of block generation.

  • When the nodes vote among each other to choose delegates, the decision of each node cannot be fully considered due to the limited voting options which cannot reflect voting in the real world. Due to the requirement of high efficiency and flexibility of nodes, a valid and elastic voting process needs to be designed.

  • Considering the human factors during the voting process, some malicious nodes may deliberately band together to cast support or against votes for some nodes due to their selfishness or own interests. Hence, how to avoid unfair voting and achieve proper delegate selection is important.

To resolve the above problems, considering the secure and effective storage [45], [46] of robotic information [38], we proposed an improved DPoS consensus mechanism based on PLTS [39] for a smart autonomous multi-robot system in this paper. The main contributions of this paper are summarized as follows.

  • Blockchain in multi-robot system: Due to the traceability and immutability of blockchain, we build a blockchain for robotic data storage in the multi-robot system for smart autonomous robots. As robotic data such as authentication, access control, configuration files and behavioral records can be stored in blockchain securely [40], [41], data cannot be tampered by remote adversaries seeking to easily break into the device for malicious manipulation [42], [43]. To accelerate block generation, nodes use DPoS to reach a consensus for generating blocks stored the robotic information.

  • Probabilistic linguistic term set is introduced to express the fuzzy decisions of the voting node in DPoS consensus mechanism in blockchain. By analyzing the fuzzy decision of the voting node, four voting options, namely “very supportive”, “supportive”, “opposed” and “very opposed”, are proposed as the specific voting decisions. Therefore, the voting options represent voting result more accurately than previous mechanisms.

  • To accelerate the generation of blocks containing robotic information, a voting algorithm with probabilistic linguistic information called VAPLI is proposed to select the delegates efficiently. This process can be formulated as a voting mechanism from a group of nodes that are encouraged to vote out the delegates to verify and produce block in DPoS consensus mechanism.

  • The score function greatly affects the voting results in VAPLI. By analyzing the proportion of options, we propose weight setting for computing the score. This process assigns different weights for various votes to avoid malicious or casual voting attitude and enhances the validity of the results. That is, if a selected option is highly valued by the voting system and has a higher reference value, then the vote of this option is not only prudently selected by the nodes in the voting process, but also occupies a larger proportion in the calculation of voting results.

  • The deviation degree is a further judgment that affects the voting results in VAPLI. The deviation degree of the score calculated by the score function can be explained by calculating the deviation. By considering the equivalent scores of nodes, a node with a greater deviation degree is more precarious and therefore not suitable to be the delegate.

The remainder of this article is organized as follows. Section 2 introduces the motivation of this paper and some basic knowledge on PLTS. In Section 3, we outline the voting algorithm with probabilistic linguistic information which is used to improve the DPoS consensus mechanism. Section 4 analyzes the improved DPoS mechanism in different voting conditions, to interpret the validity and flexibility of the proposed mechanism. Section 5 consists of experimental simulations of the improved DPoS mechanism, and the applicability and effectiveness of the consensus mechanism are verified by examples. Finally, we conclude this article and describe planned future work in Section 6.

Section snippets

Preliminaries

This section introduces the motivation of this paper and reviews of the PLTS.

Improved DPoS consensus mechanism

Due to the imprecision of voting options and the imprecise voting calculation, the most suitable delegates cannot be selected for block generation. In this paper, VALTI is developed to be more efficient, flexible and accurate in selecting delegates of DPoS.

VAPLI includes three components. The first is the normalization of the PLTS, and we can obtain a set of probabilistic linguistic terms of each voting node. The second is the calculation of an improved score function used to calculate the

Mode analysis

In this section, we first enumerate the various voting conditions of a specific voting case, and calculate the scores and probabilities of various voting conditions. Therefore, the improved DPoS consensus mechanism proposed in the voting process of n nodes is also applicable.

A case study

We assume that 4 nodes are selected as delegates from 13 nodes, which means n = 13 and m = 4. In addition, the nodes have the equal right to vote “very supportive vote”, “supportive vote”, “opposed vote” and “very opposed vote” for each other.

Table 10 shows partial voting results of 13 simulated nodes, where Ni denotes the ith node (see Table 11).

LetNiNj=2verysupportvotenodeNjtonodeNi1supportvotenodeNjtonodeNi0absentvotenodeNjtonodeNi-1opposevotenodeNjtonodeNi-2veryopposedvotenodeNjtonodeNi,i,j=

Conclusion

In this paper, by storing the relevant information of the smart robot in blockchain, the immutability of the smart autonomous multi-robot information is guaranteed, thus improving the security of the smart autonomous multi-robot system. We proposed an improved DPoS consensus mechanism based on the PLTS, and proposed VAPLI to calculate the score and deviation degree. Compared to the traditional voting system in DPoS, we establish four voting options to select the delegate node to enhance the

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work is supported by the National Natural Science Foundation (Grant No. 61772099, 61772098, 61802039); the Science and Technology Innovation Leadership Support Program of Chongqing (Grant No. CSTCCXLJRC201917); the University Outstanding Achievements Transformation Funding Project of Chongqing (Grant No. KJZH17116); the Innovation and Entrepreneurship Demonstration Team Cultivation Plan of Chongqing (Grant No. CSTC2017kjrc-cxcytd0063); the Technology Innovation and Application Development

References (49)

  • Y. Zhan et al.

    DRBFT: Delegated randomization Byzantine fault tolerance consensus protocol for blockchains

    Inf. Sci.

    (2021)
  • H. Tang et al.

    Public blockchain evaluation using entropy and TOPSIS

    Expert Syst. App.

    (2019)
  • Q.i. Pang et al.

    Probabilistic linguistic term sets in multi-attribute group decision making

    Inf. Sci.

    (2016)
  • N. Islam et al.

    A framework for topological based map building: A solution to autonomous robot navigation in smart cities

    Future Gen. Comput. Syst.

    (2020)
  • Y. Liu et al.

    Analyzing the robotic behavior in a smart city with deep enforcement and imitation learning using IoRT

    Comput. Commun.

    (2020)
  • Y. Zhang et al.

    Exploring the cognitive process for service task in smart home: A robot service mechanism

    Future Gen. Comput. Syst.

    (2020)
  • F. Herrera et al.

    A sequential selection process in group decision making with a linguistic assessment approach

    Inf. Sci.

    (1995)
  • M. Zhu et al.

    Public vehicles for future urban transportation

    IEEE Trans. Intell. Transp. Syst.

    (2016)
  • W. Dai et al.

    Cloud infrastructure resource allocation for big data applications

    IEEE Trans. Big Data

    (2018)
  • Y. Rizk et al.

    Cooperative heterogeneous multirobot systems: A survey

    ACM Comput. Surv.

    (2019)
  • W. Yanez et al.

    Data allocation mechanism for Internet-of-Things systems with blockchain

    IEEE Internet Things J.

    (2020)
  • S. Zhang et al.

    A group signature and authentication scheme for blockchain-based mobile-edge computing

    IEEE Internet Things J.

    (2020)
  • I. Vakilinia, S. Badsha, E. Arslan, S. Sengupta, Pooling approach for task allocation in the blockchain based...
  • G. Wood

    Ethereum: A secure decentralised generalised transaction ledger

    Ethereum Project Yellow Paper

    (2014)
  • Cited by (32)

    • CE-PBFT: A high availability consensus algorithm for large-scale consortium blockchain

      2024, Journal of King Saud University - Computer and Information Sciences
    View all citing articles on Scopus
    View full text