A Review on Path Planning and Obstacle Avoidance Algorithms for Autonomous Mobile Robots

Table 1

Summary of a few works performed using the classical approach.

Author	Approach	Objective	Contribution
Amaliyah et al. [11]	Dijkstra algorithm	To find the shortest distance between cities on the island of java.	The Dijkstra algorithm finds the shortest path between 46 cities by removing the node closest to the start node in Java. When utilizing Google Map as a reference, the accuracy of node combination is 92.88 percent, according to an experimental run-through.
Fuhou and Jiping [13]	Dijkstra algorithm	The authors presented a shortest path algorithm for massive data.	The authors demonstrated that DA might save a lot of memory and be used in a network with many nodes.
Cheng and Zelinsky [18]	Artificial potential field	The authors presented temporary high magnitude attraction forces at a random place to prevent the robot from getting trapped in local minima.	A theoretical analysis of a behavior-based navigation system for autonomous mobile robots is provided in this study. In an unfamiliar environment, the robot navigates and avoids obstacles.
Saravana et al. [19]	Artificial potential field	To solve obstacle avoidance problems.	An arc on a semicircle around an AUV’s bow with a predetermined radius is discretized into equiangular points, with the center representing the current position. Thus, by calculating that point, the vehicle may be steered in 2D space towards the spot with the lowest potential.
Wein, et al. [28]	Probabilistic roadmap	The authors combined the Voronoi diagram and visibility graph to get the optimal path.	The visibility–Voronoi diagram is presented as a new form of a graph. A natural-looking route is short, smooth, and maintains a certain level of consent from obstructions. In addition, we present a method for preparing a scene with configuration-space polygonal obstacles.
Yuandong and Oliver [29]	Probabilistic roadmap	An elastic RM was presented for autonomous mobile robot motion planning.	Unique feedback is offered on motion planning technique that is capable of meeting all these motion restrictions and feedback needs. This framework has been validated through simulation and real-world trials with a mobile manipulator and stationary platform.
Lingelbach [37]	Cell decomposition	The PCD was presented for a mobile robot path planning that brought milk from the fridge to the kitchen table and provided a comparison study between RRT.	The suggested method has been used to handle rigid body motions, maze-like challenges and path planning difficulties for robotic platforms.
Seda [34]	Cell decomposition	The CD was compared with the roadmap and showed that RM could purge CD disadvantages.	The disadvantages of roadmap techniques for finding a solution with polynomial time is discussed.