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

Abstract

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.

International Journal of Electronics and Telecommunications

“…The third defined trajectory T 3 is represented by matrix of poses P 3 (6) and time vector τ 3 (7) and Figure 10 shows the time plots of the desired spatial coordinates of the drone for this trajectory.…”

“…To move efficiently in the environment, the robot should have capability of creating a map of the surrounding space and locating itself in that space. These issues are very important research areas [4], [7]. The development of mobile robots is strictly connected with the invention of new materials, power systems, increasing the computing power of electronic systems and development of control algorithms.…”

A Region-Based Approach to Monocular Mapless Navigation Using Deep Learning Techniques

“…Surveys on navigation systems for robots describing map-based and mapless methods were presented in [7,8,9]. Map-based navigation systems depend on geometric models or topological maps of the environment [10]. In this case, the robot has a model of its surroundings and can detect collisions by calculating distances.…”

Motion path planning of soccer training auxiliary robot based on genetic algorithm in fixed-point rotation environment

“…The soccer robot system is a real-time dynamic and uncertain complex environment (Ying and Fei 2018;Wang et al 2018). It is a good experimental platform for studying path planning in complex environments, and it is also the focus and difficulty of soccer robot researchers (Bai et al 2018;Madhevan and Sreekumar 2018). Planning problems originate from the traditional research topics of computer geometry.…”