This article proposes an event-triggered disturbance rejection control scheme for a brain-actuated two-legged wheeled mobile robot (WMR) by using sliding mode approach. An eight-channel brain-computer interface system is developed to evoke five modes of the steady-state visual evoked potential, which are applied to control the WMR. In order to improve the speed tracking accuracy of the robot, a disturbance observer is integrated into the sliding mode controller design to actively compensate for unknown disturbances. Meanwhile, the energy usage of controlling robot is saved via introducing an event-triggering mechanism, which is designed via reachability of sliding mode. It is further shown that the tracking error can be driven into a neighborhood of the origin by the proposed sliding mode controller and event generator. The tradeoff between the convergence of sliding mode dynamics and the scheduling performance of the event generator is formulated into a multiobjective optimization problem, whose solutions are searched readily via invoking a sparrow search algorithm (SSA). Finally, the effectiveness and robustness of the proposed SSA-optimized sliding mode controller are evaluated in both human-in-the-loop simulation and experiment results. Specifically, the results demonstrate that by means of the proposed event-triggered sliding mode controller, the five subjects can complete the given tasks successfully with maintaining the disturbance rejection performance and saving 45.90% +/- 2.02% usage of computation resources for the brain-actuated WMR.