This paper studies the consensus problem for a class of nonlinear multi-agent systems with asymmetric time-varying output constraints and completely unknown non-identical control directions. Firstly, in order to deal with the problem of asymmetric time-varying output constraints, the original output-constrained multi-agent systems are transformed into new unconstrained multi-agent systems by constructing the state transformation for each agent. Secondly, the emergence of multiple Nussbaum-type function terms is avoided by introducing novel sliding-mode-esque auxiliary variables and consensus estimate variables, which allows the control directions to be completely unknown non-identical. Thirdly, a novel control strategy is proposed by combining novel variables with state transformation method for the first time, which makes the design of distributed consensus protocol more concise. Through Lyapunov stability analysis, the proposed distributed protocol ensures that the output constraints are never violated and the consensus can be achieved asymptotically. Finally, a practical simulation example is given to demonstrate the effectiveness of the proposed distributed consensus protocol.

本文研究一类具有非对称时变输出约束和完全未知的非相同控制方向的非线性多智能体系统的一致性问题。首先，为了处理非对称时变输出约束问题，通过构造每个智能体的状态变换，将原来的输出约束多智能体系统转化为新的无约束多智能体系统。其次，通过引入新颖的滑模式辅助变量和共识估计变量，避免了多个Nussbaum型函数项的出现，使得控制方向完全未知不相同。第三，首次将新变量与状态转换方法相结合，提出了一种新的控制策略，这使得分布式共识协议的设计更加简洁。通过李雅普诺夫稳定性分析，所提出的分布式协议确保永远不会违反输出约束，并且可以渐近地达成共识。最后，给出了一个实际的仿真例子来证明所提出的分布式共识协议的有效性。