Underwater glider is one of the most promising autonomous observation platforms for long-term ocean observation, which can glide through water columns by adjusting its buoyancy and attitude. In this paper, gliding range of the Petrel-L underwater glider is increased by optimizing the motion parameters according to different observation missions. The optimization is established to maximize the gliding range and minimize the energy consumption for one gliding cycle, which is solved by the inner penalty function method (IPFM). Based on the nonlinear variation rule of the buoyancy loss, a hidden gliding strategy model (HGSM) is proposed and the optimization result indicates that it can increase the gliding range by 24.13% for Petrel-L when the power related with time is 0.2 W. The optimization is applicable to other types of underwater gliders.