<p>This study investigates the adaptive dynamic event-triggered tracking control problem for flexible-joint robots which operate amidst random disturbances and have unknown measurement sensitivity. The systems considered with stochastic disturbance terms, belong to the category of coupled multi-input multi-output stochastic nonstrict-feedback systems, thereby heightening the complexities of control. Moreover, in the case of unknown measurement sensitivity, the actual states cannot be obtained anymore, which also increases the difficulty of control. Hence, fuzzy logic systems occupy a significant role in this study, which not only approximate the unknown terms arising from random disturbances, unknown measurement sensitivities and system uncertainties, but also avoids the algebraic loop problem. Based on the backstepping method, an adaptive dynamic event-triggered controller is designed, the proposed control method can not only address the impact of random disturbances, but also enable the real tracking errors to converge to a small neighborhood even when the measurement sensitivity is unknown. Rigorous stability analysis conclusively establishes all signals of the closed-loop system are bounded in probability, while a proof by contradiction effectively demonstrates the absence of Zeno behavior. Ultimately, the efficacy of our designed control strategy is confirmed through the rigorous simulations. In the comparison simulation, it can be observed that our tracking error remains relatively small even in the presence of unknown measurement sensitivity.</p>