论文部分内容阅读
多智能体一致性协调控制的最终收敛状态受限于通信拓扑结构与边的权值,而收敛状态的不同进一步影响多智能体趋同的速度.为实现拓扑结构与协调收敛状态解耦,保证最短时间实现一致性,本文设计一种输入受限线性多智能体分布式协调控制策略.首先基于Helly定理证明了n个输入受限线性多智能体系统在d(n>d)维协调空间上的最短时间一致性协调状态和收敛时间唯一存在,并取决于其中至多d+1个智能体.当找到该d+1个起决定作用的智能体后,即可得到所有智能体的最短时间一致性状态.根据此定理,设计一种新的分布式协调算法使得各个智能体知道起决定作用的智能体,进而计算得到协调收敛状态与收敛时间,随后各个智能体独立设计含终端时间和终端状态约束的局部最优控制律,保证最短时间一致性实现.最后在二阶线性多智能体系统上进行仿真验证.仿真结果验证了分布式算法的可行性,并且当协调状态维度远小于智能体数量时,计算量明显减少,计算速度显著增加.
The final convergence of multi-agent coherency control is limited by the communication topology and the weights of the edges, and the different convergence states further affect the convergence speed of multi-agent.In order to decouple the topological structure and coordination and convergence, the shortest guarantee Time consistency, we design a distributed and coordinated control strategy based on input limited linear multi-agent.Firstly, based on the Helly theorem, we prove that n input-constrained linear multi-agent systems in d (n> d) The shortest time consistency coordination state and the convergence time exist only and depend on at most d + 1 agents. When the d + 1 determined agents are found, the shortest time consistency of all agents can be obtained According to this theorem, a new distributed coordination algorithm is designed so that each agent knows the determinant agent, then the coordination and convergence time can be calculated and the convergence time can be calculated. Then each agent can be independently designed with constraints of terminal time and terminal state The local optimal control law is guaranteed to ensure the shortest time consistency.Finally, simulation is carried out on the second-order linear multi-agent system. The feasibility of the distributed algorithm, and when the state of the coordination number of dimensions much smaller than smart body, significantly reduced amount of calculation, a significant increase in computational speed.