Nowadays, higher requirements are placed on the design of the controller when robots enter the factory and co-operate with human. In this letter, we propose a unified task space control method combined with the contact reaction strategy in null space of redundant manipulator, which can not only conveniently switch between different control objectives according to specific tasks, but also react to the unknown contact with compliance. Specifically, the unknown contact force applied on the manipulator’s body is online estimated based on recursive least square estimation law in joint space. Then, the closed-loop dynamics of different control objectives are unified in task space by a generalized target model, and a model-reaching law based on adaptive robust control is proposed to achieve the target model even if there exists an unknown contact. Meanwhile, the estimated contact force is not only compensated in the model-reaching control law to guarantee the correct task execution but also used for the null space impedance control design to ensure compliant behavior. The performance and stability of the proposed scheme are guaranteed and proved in theory. Experiments are conducted on a 7-DoF manipulator for the validation.