In bio-inspired robotics, use of a Central Pattern Generator (CPG) to coordinate actuation is fairly common. The gait achieved depends on a number of CPG parameters, which can be adjusted to control the robot's motion. This paper presents an output feedback motion control framework, addressing issues encountered when dealing with this type of control problem, including partial state measurements and system uncertainty. Efficacy of the presented approach is illustrated by results of numerical simulations in the case of a swimming robot.