A universal error transformation method is proposed for designing a distributed event-triggered formation tracker with preserved network connectivity and collision avoidance for multiple pure-feedback nonlinear multiagent systems connected in a directed network. It is assumed that the communication and avoidance ranges of agents are heterogeneous and that all nonaffine nonlinear functions are unknown. The primary contribution of this study is to establish a universal nonlinear transformation of relative output errors among agents for dealing with the initial network interaction, collision prevention, and predefined-time tracking performance problems in a fully distributed manner. To this end, nonlinear relative output errors among agents are designed using performance functions dependent on communication and avoidance ranges. Local event-triggered adaptive control laws are recursively designed using nonlinear relative output errors and neural networks. It is proved that the boundedness of the presented nonlinear errors ensures the achievement of network interaction preservation, collision prevention, and formation tracking with predefined-time convergence. Furthermore, the existence of a minimum inter-event time is established for local event-triggered control laws. Finally, the effectiveness of the proposed theoretical strategy is validated by simulation examples.

针对连接在有向网络中的多个纯反馈非线性多智能体系统，提出了一种通用误差变换方法，用于设计具有保留网络连通性和避免冲突的分布式事件触发编队跟踪器。假设代理的通信和回避范围是异构的，并且所有非仿射非线性函数都是未知的。本研究的主要贡献是建立代理之间相对输出误差的通用非线性变换，以完全分布式的方式处理初始网络交互、碰撞预防和预定义时间跟踪性能问题。为此，使用依赖于通信和回避范围的性能函数来设计代理之间的非线性相对输出误差。使用非线性相对输出误差和神经网络递归地设计局部事件触发的自适应控制律。证明了所提出的非线性误差的有界性确保了网络交互保持、碰撞预防和具有预定义时间收敛的编队跟踪的实现。此外，为局部事件触发控制律建立了最小事件间时间的存在。最后，通过仿真实例验证了所提理论策略的有效性。和具有预定义时间收敛的编队跟踪。此外，为局部事件触发控制律建立了最小事件间时间的存在。最后，通过仿真实例验证了所提理论策略的有效性。和具有预定义时间收敛的编队跟踪。此外，为局部事件触发控制律建立了最小事件间时间的存在。最后，通过仿真实例验证了所提理论策略的有效性。