This paper provides a new approach to the dynamic motion planning problems of mobile robots in uncertain dynamic environments based on the behavior dynamics from a control point of view. The fundamental behavior of a mobile robot in motion planning is regarded as a dynamic process of the interaction between the robot and its local environment, and then it is modeled and controlled for the motion-planning purpose. Based on behavior dynamics, the dynamic motion-planning problem of mobile robots is transformed into a control problem of the integrated planning-and-control system. And the dynamic motionplanning problem can be transformed into a conventional optimization problem in the robot's acceleration space. Realization of the collision-avoidance behavior is shown to be just a control problem of the robot's acceleration. The proposed method can directly provide the desired acceleration for mobile robots. No restrictions are assumed on the shape and trajectories of obstacles. No local minima are encountered in most cases. Collision avoidance between multiple mobile robots can also be realized. Stability of the whole planning-and-control system can be guaranteed. Our method provides a new insight to the motion-planning problem of mobile robots based on behavior dynamics and from the control point of view. Simulation experiments illustrate our results.