Inertial sensors are key components of gravitational wave observations and Earth geodesy missions. An inertial sensor includes an isolated free-floating test mass (TM) surrounded by capacitive electrodes and a housing frame (EH) to perform the relative-position measurement and control the TM in six degrees of freedom. Owing to galactic cosmic rays (GCRs) and solar energetic particles (SEPs), many additional accelerations are introduced through the Coulomb interaction between charged TMs and their surrounding conducting surfaces. Thus, the TM charge control is critical in space-based missions. A contact-free and ultraviolet (UV) light-based charge management system (CMS) was developed to reduce charge-induced noises acting on the TMs and minimize force disturbances that can perturb measurements or interrupt science tasks. However, the operating environment for space charge control is full of uncertainties and disturbances. Physical parameters in the discharging process are rarely measured and will vary owing to changes in solar activity, temperature, and so on. The unpredictability and variability of these parameters affects the CMS performance in long-term space missions and must be evaluated or eliminated. This paper presents a simplified physical model for the discharge process based on electron exchange between the TM and the opposing EH. Subsequently, a model reference adaptive control (MRAC) is proposed for the CMS with parametric uncertainties to maintain the TM charge below a certain level and improve its robustness. The simulation results show that the MRAC can automatically adjust control parameters to eliminate the effect of the variability of the aforementioned physical parameters, and the control precision can reach 0.1 mV under uncertainties, which is superior to that of a classic proportional-integral-derivative controller. This study demonstrated the effects of adaptive charge control and its potential for actual applications.