Strategic Creation and Placement of Landmarks for Robot Navigation in a Partially-known Environment

Abstract

Navigation of autonomous robots in uncharted or uncertain, and constrained or hazardous environment is a common problem but integrated with multiple challenges. This paper considers a subsection of such an environment and essays a feasible yet innovative solution for accurate parking and obstacle avoidance of a car-like mobile robot. The new method is based on strategic creation and positioning of landmarks in a bounded workspace that will aid or guide the robot to safely navigate the workspace and finally park correctly inside of the designated parking bay. By autonomously controlling its translational velocity and the steering angle, the car-like robot navigates from one (newly fixed) landmark to another (newly fixed) and finally converges to a target with a pre-defined posture. Attaining accurate posture is very important in real-life situations which involve tasks such as loading or off-loading and deliveries in constrained spaces. The paper establishes practical posture stability of the system. The computer simulations verify the effectiveness of the proposed method and the proposed control inputs.