Moving Obstacle Avoidance in Indoor Environments

This paper deals with a problem of autonomous mobile robot navigation among dynamic obstacles. The velocity obstacle approach (VO) is considered an easy and simple method to detect the collision situation between two circular-shaped objects using the collision cone principle. The VO approach has two challenges when applied in indoor environments. The first challenge is that in the real world, not all obstacles have circular shapes. The second challenge is that the mobile robot cannot sometimes move to its goal because all its velocities to the goal are located within collision cones. For indoor environments some unobserved moving objects may appear suddenly in the robot path; particularly when the robot crosses a corridor or passes an open door. The contributions of this paper are that a method has been proposed to extract the collision cones of non-circular objects, where the obstacle size and the collision time are considered to weigh the velocities of the robot, and the virtual obstacle principle has been proposed to avoid unobserved moving objects. The experiments were conducted within indoor environments to validate the control algorithm proposed in this paper, and results obtained showed the mobile robot successfully avoided both static and dynamic obstacles.