To address the safety interactive control problem for robotic manipulator in disturbed environments, a whole-body dynamic collision avoidance and safety compliance control method based on a control barrier function is proposed. To ensure the robust safety while reducing conservatism for improved tracking performance, a nonlinear disturbance observer-based tunable input-to-state safe high-order control barrier function (NDOB-TISSf-HoCBF) is proposed, which can handle high relative degree constraints, and it extends the applicability of the input-to-state safety control barrier function. Then, inequality constraints are established using the proposed control barrier function, and a robust safety critical controller based on the NDOB-TISSf-HoCBF is formulated by a quadratic program (QP) that modifies a nominal controller with minimal intervention. Virtual safety barriers are established between the obstacles and multiple positions of the links of the robotic manipulator, in order to achieve the avoidance of static/dynamic obstacles by the entire robot manipulator. At the same time, multiple dynamic spatial constraints are established at the end of the robotic manipulator to ensure that the manipulator performs interative tasks in the preset safety area. In addition, the forward invariance of the safe set is proven using the Nagumo's theorem, and the effectiveness of the proposed control method is validated through simulations and experiments.