This study presents an adaptive fuzzy broad-learning neural control (AFBNC) system applied to a reaction wheel pendulum without knowing its dynamic model. The AFBNC system uses a fuzzy broad-learning system (FBLS) to approximate an ideal controller online with keeping the reaction wheel pendulum balance. Moreover, the gradient descent method and chain rule are applied to online adjust all parameters of the FBLS with keeping the closed-loop control system stable. Finally, to implement the control algorithms and conduct experiments, a prototype of the reaction wheel pendulum is designed using a low-cost microcontroller. The experimental results demonstrate the effectiveness of the proposed AFBNC system using low-cost hardware.