Abstract: The paper presents a model of cerebellar control that can learn to produce motor signals in a muscle synergy space for the purposes of reaching a specified target.The model consists of a learning celrebellar module and a hard-wired extra-cerebellar (EC) system.The model is used to control a simulated two-joint arm,which is actuated by a set of six muscles.In the model,the cerebellar module consists of a group inde- pendent submodules,referred to Adjustable Pattern Generators(APG);and the EC system is responsible for producing low-quality corrective movements in situations where the cerebellar module is unable to bring the arm to the specified target.Via proprioceptive inputs,the inferior olive of the model assesses directional error in cerebellar-generated movements by observing subsequent corrective movements produced by the EC system. The simulation results demonstrate that this cerebellar model has better tracing performance of desired trajectory.