<p>This paper proposes an adaptive fuzzy backstepping optimal control method of modular manipulators based on self-triggering. Through the application of joint torque feedback approaches, a dynamic model of the system is established, and state space characterization is derived. Based on the joint angular velocity error and position error contained in the fusion function, the cost function is established. The unknown nonlinear dynamics in the system are approximated online through the fuzzy logic system (FLS), and a virtual control law is designed in combination with backstepping control to solve the problems of model uncertainty and high-order nonlinearity. Self-triggering mechanism has been introduced, with the system state at each triggering moment being used to predict the next trigger instant and update the control strategy, thus avoiding continuous monitoring of the system state under event-triggered conditions and reducing the resource waste. According to Lyapunov’s stability theory, the developed approach guarantees uniform ultimate boundedness stability for the closed-loop system while effectively preventing Zeno behavior. Finally, experiments are conducted on a two-degree-of-freedom modular manipulators platform, and the effectiveness of the control method is confirmed.</p>