CAJ | 학술논문

传统基于避撞、组队和聚集规则的个体运动协同算法具有内聚和速度一致趋势,群体在外部信息刺激下难以自发实施分群.为此,提出一种融合了邻域跟随行为的分布式协同控制算法.该算法在短距排斥、长距吸引和速度一致行为的基础上,引入个体对于其感知域内间距变化最快的邻居的跟随运动,并通过对跟随目标的动态更新,实现了外部信息作用下群体的自组织分群行为.仿真实验验证了算法的可行性和分群有效性.
전통기우피당、조대화취집규칙적개체운동협동산법구유내취화속도일치추세,군체재외부신식자격하난이자발실시분군.위차,제출일충융합료린역근수행위적분포식협동공제산법.해산법재단거배척、장거흡인화속도일치행위적기출상,인입개체대우기감지역내간거변화최쾌적린거적근수운동,병통과대근수목표적동태경신,실현료외부신식작용하군체적자조직분군행위.방진실험험증료산법적가행성화분군유효성.
Traditionally, the motion cooperative algorithms for swarm base on the interaction rules of attraction, repulsion and alignment have the properties of group cohesion and velocity consensus, which prevent the splitting of the group under multiple external stimulus. Therefore, a distributed motion cooperative algorithm is proposed by integrating a following interaction with the behaviors of long-range attraction, short-range repulsion and consensus-based alignment, where the following behavior is applied locally to one of the proximity neighbors with the fastest change of inter-individuals distance and vanished after a short acting time. Then, the swarm can split autonomously into multiple sub-groups when multiple stimulus are introduced. Finally, simulation results show the feasibility and effectiveness of the proposed algorithm for the fission control.