Research On The Self-Organized Fission Control Of Flocking System Based On Information Coupling Degree

The fission behavior of flocking system is the spontaneous splitting phenomenon from a cohesive flock to multiple sub-clusters.As the inherent motion pattern of flocking system,fission behavior not only remedies the defect that the single fusion behavior is unable to characterize the splitting phenomenon of a flock,but also en-dues the flock more flexibility of movement and diversity of behavior,which is the typical manifestation of environmental adaptivity.Therefore,investigation on the fis-sion control of flocking system is of crucial theoretical value and practical meaning in revealing the internal mechanism of the "fission-fusion" behavior of flocking sys-tem,deepening the cognition of the mechanism of self-organized collective motion and promoting the engineering application of fission behavior.Supported by the National Natural Science Foundation of China(No.51179156),this dissertation addresses the self-organized fission behavior without any intelligent planning methods such as ap-pointment or negotiation.By a detailed analysis on the internal mechanism of fission behavior,an in-depth study on the self-organized fission control method and related problems for flocking system based on information coupling degree is provided.The main contributions and research achievements of this dissertation are as following:(1)Research on the biological mechanism and modeling of self-organized fission behavior for flocking systemTaking the fission behavior of biological flocks as the reference,the information transfer principle among members in the flock is investigated from a directional in-formation propagation perspective,which demonstrates that the fission behavior is virtually an individual motion differentiation phenomenon caused by the directional information transfer of multiple conflicting external stimuli and the correlation dif-ference between members is the primary cause for the fission behavior.Accordingly,information coupling degree is defined to denote the correlations between individuals,and an information coupling degree based fission control framework is proposed,which lay the foundation for the design of the subsequent fission control algorithm.(2)Research on the fission control method of flocking system based on information coupling degreeInspired by the interaction relationship of individuals in biological flocks,the model of information coupling degree is constructed by synthesizing both position and velocity information.Then,a "max-ICD" fission control strategy is proposed from the maximum stimulus information transfer perspective.By integrating the "pairwise interaction" rule into the traditional "separation/alignment/cohesion" rules,an infor-mation coupling degree based fission control algorithm is designed,which realizes the self-organized fission behavior of flocking system under external stimuli.Both theo-retical analysis and simulation experiments validate the effectiveness of the proposed fission control algorithm.(3)Research on the fission control method of flocking system without velocity measurementIn consideration of the absence of velocity measurement in the fission control of flocking system,a fission control method with position information only is pro-posed.This method utilizes the information coupling degree based fission control framework as its foundation.In consideration of the absence of velocity measurement,a distributed observer is built to estimate the relative velocity information from their relative positions.In addition,the relative position and estimated relative velocity are deployed to design the fission control algorithm,which realizes the self-organized fission behavior without velocity measurement.Both theoretical analysis and simu-lation experiments validate the effectiveness of the proposed fission control algorithm without velocity measurement.(4)Research on the fission control method of flocking system with time delayFor the time delay problem in the information acquisition process of fission con-trol,the limitations of the information coupling degree based fission control algo-rithm in the presence of time delay is firstly investigated,which demonstrates that the existence of time delay will significantly reduce the performance of fission control algorithm.Then,the fission control problem with time delay is studied from the per-spective of constant time delay and time-varying time delay.For the case of constant time delay,Lyapunov-Razumikhin theorem is adopted to derive the sufficient condi-tion as well as the allowable time delay upper bound for the stable fission behavior.For the case of time-varying time delay,Lyaponov-Krasovskii functional is deployed to obtain the time delay upper bound and the convergence condition in the form of LMIs.Both theoretical analysis and simulation experiments validate the effectiveness of the proposed fission control algorithm with time delay.(5)Research on the experimental validation of the fission control for swarm robot systemTo verify the effectiveness of the proposed information coupling degree based fission control method in the application of real flocking system,experimental vali-dation is performed on the E-puck swarm robot experiment platform.According to the characteristics of E-puck robot,the kinematic model is established.Together with the measured range-bearing information of neighbors,the fission control algorithm for swarm robot system is designed under the information coupling degree based fission control framework.Simulation on the Webots platform illustrates the feasibility and effectiveness of the proposed algorithm.