A Simple Limit Cycle Negotiation Strategy for Robot Navigation

Limit cycles can occur when navigating unmanned ground vehicles (UGVs) using behavior-based algorithms. Limit cycles occur when the robot is navigating towards the goal but enters an enclosure that has its opening in a direction opposite to the goal. The robot then becomes effectively trapped in the enclosure. This paper reports a solution to the limit cycle problem for robot navigation in very cluttered environments, for example dense forests. These types of environments offer a challenge due to the diversity of shapes and sizes of deadlocks that are likely to appear. A simple deliberative algorithm for detecting and retracting from limit cycles is described. The algorithm uses spatial memory to detect the limit cycle. Once the limit cycle has been detected, a labeling operator is applied to a local map so that the obstacles that form the boundary of the deadlock enclosure are identified. Subsequently, the robot is directed outside the enclosure using a behavior based control system. Once it exits this region, the deadlocked area is designated as off-limits by means of a virtual wall. Finally, the robotic vehicle proceeds to its original target avoiding the virtual wall and the different obstacles that are found on its way. Simulation and experimental results demonstrate the effectiveness of the proposed method.