Dead-lock free mobile robot navigation using modified artificial potential field

This paper presents a dead-lock free motion planning algorithm for mobile robot navigation in an unknown environment. The proposed algorithm uses the Gaussian function based artificial potential field (APF). Here, we present an improved APF based motion planning algorithm for omnidirectional mobile robot which searches the goal point in an unknown 2D environment. APF algorithm is modified to eliminate dead-lock and non-reachability problems of navigation. This method proposes a new approach to escape the robot from the local minima where the dead-lock happens. We have given effective front-face obstacle information associated with the velocity direction into the potential field method. This modification with additional information solves the dead-lock problem that the APF often converges to local minima. Mathematical model of the new approach is explained and simulation experiments performed in the MATLAB™ environment verifies that the proposed method is correct and effective for dead-lock free motion of mobile robots.