Cycling induced by functional electrical stimulation (FES) coupled with motorized assistance is a promising rehabilitative strategy. A switching controller that activates lower limb muscles alongside an electric motor based on the crank angle is developed to facilitate cycling. Due to the periodic nature of cadence tracking in cycling, a repetitive learning controller (RLC) is developed to track a desired cadence trajectory with a known period. The RLC is developed for an uncertain, nonlinear cycle-rider system with autonomous state-dependent switching. Electrical stimulation switches across multiple lower limb muscle groups based on the torque effectiveness throughout the crank cycle. The electric motor provides assistance when the muscle groups yield low torque production. A Lyapunov-based stability analysis that invokes a recently developed LaSalle-Yoshizawa corollary for nonsmooth systems is used to guarantee asymptotic tracking. The developed controller was tested during FES-cycling experiments in five able-bodied individuals and three participants with neurological conditions. The added value of the RLC in cadence tracking is illustrated by comparing the results of two trials with and without the learning feedforward term. The results indicate that the RLC yields a lower mean root-mean-squared cadence tracking error.