Secure Control Architecture And Algorithm Based On Blockchain Towards Swarm Robotics' Multi-decision Making Scenario

The swarm robot technology is different from the single robot technology which means a brand new type of robot's technique that uses multiple robots,autonomous networks and decentralized control to manage and control.The origin of the content related to its control algorithms,communication links and task management from the observation and study of colony animals in the last century.It has been proven that this type of technology has broad application prospects in the future,especially in many areas where single-body robots cannot be perfectly competent.Existing swarm robots' researchers focus on the control and maintenance of swarm state,data exploration and network formation.There has not been so much research on data security issues.Blockchain,as a new type of data storage technology,has proven to be of great applied value in terms of resistance to DDoS attacks,data confidentiality and integrity.The prominent characteristics of its decentralization coincide with the swarm robots,but there are few studies on the combination of these twos.In this thesis,we aimed at swarm robots in the field of exploration as a service and study two major issues which are exploring malicious attackers in the process of service outsourcing and exploring data security issue at the executing state.Based on this,the author proposes a complete set of solutions to the above problems based on blockchain technology.The specific content includes the service layer architecture design of the service outsourcing process' s security issues and the execution layer architecture design of the execution level's security issues.Besides the architecture itself,this thesis includes the in-depth and detailed description for algorithms on security control process and blockchain consensus,and finally conducts a security analysis for the two-layer architecture.Depicted by the result,our design can resist general attacks preferably.We believe that the architecture for execution layer needs to be pluggable,replaceable,task-specific architecture,and cannot be designed into a general execution architecture.So we choose multi-objective decision-making tasks as a background for architecture design beyond execution layer.The key to the application of blockchain technology to swarm robots is that they need to be able to support the mobility of swarm robots,the grouping of networks,and the efficiency of data.Based on the above problems,this thesis designs two experiments at the end to explore the efficiency problem in the grouping process and the practicality problem of the new segmentation tolerance consensus algorithm.