Identification of probabilistic approaches and map-based navigation in motion planning for mobile robots

Robot motion planning (RMP) develops a precise path between start and goal points for mobile robots in an unknown environment. RMP is a complex task when it needs to be planned for a group of robots in a coordinated environment with leader–follower relationship. The planned path might change depending upon the number of robots and the decision made. The decision made by each robot depends on the feedback received based on the subsequent action taken by other robots. In addition, the computational complexities depend upon factors such as communication between robots, the influence of moving obstacles and environment in which they are interacting. In order to explore further in the area of motion planning, it is felt that a comprehensive survey of available literature would support researchers working in RMP and hence the present paper. This paper reviews around 152 articles published in various international journals and conferences with more emphasis on articles published after 1960. In this work, recent activities carried out in the field of path planning for mobile robotics are critically evaluated and problems faced by the researchers are also highlighted. The focus is towards implementation of probabilistic algorithms, including Probabilistic Road Map and Simultaneous Localization and Mapping. Future research prospects in multi-robot path planning based on probabilistic approaches are also discussed.