A review of navigation techniques for telepresence robots

Abstract

Nowadays the need for digital travel has increased more than ever. This is due to the need for social distancing for medical reasons, economy of travelling, and avoidance of risks involved in physical travelling or for convenience. The telepresence robot therefore serves as a surrogate for the pilot user. Telepresence robots are very resourceful and versatile. Their applications include education; medicine; business; research and ambient assisted living. A major consideration for the effective functioning of telepresence robots is the navigation scheme employed which is central to its overall reliability. Depending on the use to which the robot is to be put, several authors have considered various navigation schemes used to effectively control and navigate in workspace consisting of obstacles and/or order robots. This work aims to review the navigation technique that has been employed by researchers for telepresence robots with a view to identifying the strengths and weaknesses of the various navigation techniques and subsequently coming up with a relatively novel system of navigation that attemts to mitigate the various weaknesses of the previous methodologies for telepresence robot navigation. It has been observed that the general path planning and navigation for the usual autonomous robot is also applicable to the telepresence robot but bearing in mind that the robot is being piloted by humans and it is only when in a condition where it would be cumbersome for the humans that autonomous control system takes over. At such times path planning and navigation using AI methods have proved to be of importance. The path planning algorithms have been shown to be categorized into two viz: global path planning algorithms which are used for known environments and local path planning algorithms for unknown or partially known environments. The algorithms for local path planning are majorly reactive in attribute and are more relevant to telepresence robots. In any particular application, an appropriate blend of a number of these algorithms is employed to achieve desired navigation objectives. Furthermore, in this work, we propose the combination of a line follower robot with an obstacle avoidance algorithm for situations where the expected robot paths are foreknown.