A Review of Path-Planning Approaches for Multiple Mobile Robots

Abstract

Numerous path-planning studies have been conducted in past decades due to the challenges of obtaining optimal solutions. This paper reviews multi-robot path-planning approaches and decision-making strategies and presents the path-planning algorithms for various types of robots, including aerial, ground, and underwater robots. The multi-robot path-planning approaches have been classified as classical approaches, heuristic algorithms, bio-inspired techniques, and artificial intelligence approaches. Bio-inspired techniques are the most employed approaches, and artificial intelligence approaches have gained more attention recently. The decision-making strategies mainly consist of centralized and decentralized approaches. The trend of the decision-making system is to move towards a decentralized planner. Finally, the new challenge in multi-robot path planning is proposed as fault tolerance, which is important for real-time operations.

Optimal fusion‐based localization method for tracking of smartphone user in tall complex buildings

“…Improved convergence WDE uses a weighting scheme to assign different importance to different particles, improving convergence towards the optimal solution DE does not use a weighting scheme and may converge more slowly or get stuck in local optima [64][65][66] More robust to noise WDE can be more robust to measurement noise due to the use of a weighting scheme and more accurate tracking of the target DE may be more sensitive to noise and may require more particles to maintain accuracy [64][65][66] Better handling of multimodal distributions WDE can better handle multimodal distributions, which can occur when tracking multiple targets or in cluttered environments DE may struggle to handle multimodal distributions and may converge towards a single mode [35,64,66] Better exploration of search space WDE can explore the search space more effectively due to the use of a weighting scheme, allowing particles to explore different regions of the space DE may be more prone to getting stuck in local optima and may require more iterations to explore the entire search space [64][65][66] Better scalability to high-dimensional spaces WDE can handle high-dimensional spaces more effectively by using a weighted scheme to assign different importance to particles in different regions of the space DE may struggle with high-dimensional spaces due to the curse of dimensionality and the need to evaluate a large number of candidate solutions [64][65][66][67] 1120 -JAMIL and KIM…”

Multi-Agent Action Graph Based Task Allocation and Path Planning Considering Changes in Environment

“…Task allocation [1] and path planning [2] are very important for multi-robots to perform tasks in a coordinated manner. Although there have been many studies on these planning problems in the past, it is difficult for these studies to plan for changes in the mobility of robots in the environment.…”

Investigation on the Mobile Wheeled Robot in Terms of Energy Consumption, Travelling Time and Path Matching Accuracy

“…The issues related to it are also extremely important in the case of electric vehicles. A large group of solutions are those based on machine learning [8,9] The article [10] presents an overview of how to approach this problem. The decision-making strategy and path-planning algorithms for various types of robots are described.…”