The application of mobile manipulators in industrial dynamic cluttered environments is becoming increasingly prevalent. Therefore, a mobile manipulator system that can effectively predict the movement trajectories of dynamic obstacles and perform obstacle avoidance planning in dynamic environments is of great significance. The RAMPAGE framework, as a newly proposed emerging dynamic obstacle avoidance planning framework, possesses efficient and reliable path planning capabilities. However, it cannot predict the movement trajectories of dynamic obstacles. Thus, in this paper, we integrate the Trajectron++ trajectory prediction model with RAMPAGE to enhance its prediction capabilities. Through simulation experiments, it has been verified that the RAMPAGE framework, after integration, gains the ability to predict dynamic obstacles. However, the planning speed and operational success rate have decreased. This integration model still represents a significant challenge in mobile robotics, combining machine learning-based prediction with robust motion planning, enabling more effective navigation in complex dynamic environments.