The obstacle avoidance is currently treated by methods that fall into two broad categories: global and local approach. This paper considers the obstacles whose velocity and direction cannot be easily predicted. Such obstacle is called transient obstacle. To avoid such kind of obstacle, we introduce a local path planning method for a robotics by using cellular automaton approach. The cellular automaton was combined with Dijkstra shortest path algorithm as global path planning to obtain a path for mobile robot to be able to avoid transient obstacle along the path. Using the proposed method, a scene in a typical corridor has been created. Moreover, this paper also evaluated two kinds of obstacle avoidance motion. First, the robot uses the “stop and go” method, which is the robot decreases its speed while encounter a transient obstacle. The second one is detour method, in which the robot makes a detour motion to avoid a transient obstacle. To coupe the drawbacks of local path planning, this paper also propose the enhancement of detour method. The simulation results show that in dynamic environment with transient obstacles, the “stop and go” method produces minimal collision with shortest-distance path. While, using the detour method generates minimal collision with time-minimal navigation path.