Page 249 - 2024-Vol20-Issue2

245 |
cell represents an obstacle or free space. The algorithm
starts from the goal position, assigns wavefront values,
and propagates them outward until it reaches the start
position, creating a path.
• Genetic Algorithms: Genetic Algorithms (GA) are
population-based optimization techniques inspired by
the process of natural selection [9]. In path-planning,
GA involves evolving a population of candidate paths
using genetic operations such as crossover and muta-
tion. Paths with higher fitness, determined by a fitness
function, have a higher chance of being selected for the
next generation, gradually improving the quality of the
paths.
• Artificial Potential Fields: Artificial Potential Fields
(APF) is a reactive path-planning approach that models
the environment as attractive and repulsive forces [10].
The robot navigates by following the gradient of the
potential field, moving towards the goal while avoiding
obstacles. APF is well suited for real-time applications,
but it may suffer from local minima and oscillations.
• Ant Colony Optimization (ACO): Ant Colony Op-
timization (ACO) is a metaheuristic inspired by the
foraging behavior of ants [11]. In path-planning, ACO
involves simulating the movement of virtual ants that
deposit pheromones on the paths they explore. The
algorithm utilizes pheromone trails to guide other ants
in finding shorter and more efficient paths to the goal.
• Harmonic Functions-Based Methods: Harmonic Func-
tions Based Methods use potential functions derived
from harmonic functions to compute optimal paths [12].
These methods represent the environment as a graph
and solve partial differential equations, ensuring that
the computed paths follow the laws of physics and offer
smooth trajectories.
In addition to the aforementioned algorithms, there are sev-
eral other path-planning techniques used in various scenarios.
For instance, Visibility Graphs are used to compute paths
in environments where direct line-of-sight visibility between
points can be guaranteed. They connect visible points with
straight line segments, creating a graph that simplifies path
planning [13]. Cell decomposition is a common technique
used to simplify the path planning process, especially in struc-
tured environments. It is used to divide the robot’s envi-
ronment into smaller, manageable regions or cells. Each
cell represents a portion of the environment, and these cells
are typically of a uniform size and shape [14]. Finally, the
Minkowski sum is used, in the context of robotics and path
planning, to calculate the space that encompasses all possible
Sabeeh & Al-Furati
positions of a robot when it moves in a particular environment
while considering its geometry and size. When taking the
Minkowski sum of the robot’s shape and the obstacles in the
environment, we can create a new shape that represents the
free space in which the robot can navigate without colliding
with obstacles. This free space is essential for path planning
algorithms to find collision-free paths for the robot [15].
B. Path-Planning in Healthcare Settings
In recent times, path planning in healthcare has seen signifi-
cant progress, with intelligent algorithms playing a vital role
in various medical applications. Healthcare facilities’ com-
plexities have led to the development of advanced path plan-
ning methods that ensure smooth navigation. These solutions,
powered by cutting-edge algorithms and technology, aim to
enhance patient care, improve robotic assistance, and opti-
mize logistical operations in the medical sector. This section
explores path planning in healthcare, emphasizing the transfor-
mative impact of AI-driven approaches on medical workflows
and seamless movement in critical healthcare environments.
For instance, in a notable study, Z. D. Hussein, M. Z.
Khalifa, and I. S. Kareem [16] improved the performance
of a Laparoscope surgical robot with seven degrees of free-
dom. They used a genetic algorithm and a MATLAB-based
program to find the best path, optimizing distance while avoid-
ing obstacles in dynamic environments. Real-world tests
were conducted at Al-Sader educational hospital and the Re-
search Unit of Automation and Robotics, University of Tech-
nology, using the Lab-Volt Servo Robot System Model5250
(RoboCIM5250). This advancement has the potential to en-
hance surgical procedures and reduce patient risks. However,
the findings’ practical relevance may vary depending on each
hospital’s robot and environment characteristics.
In another recent study [17], Fang et al. introduced ad-
vanced technology called SLAM (Simultaneous Localization
and Mapping) for robots in healthcare facilities. This tech uses
images to make hospitals run smoother and reduce COVID-19
risks. It is good at handling moving things in hospitals by
understanding pictures. Combining this with a knowledge
graph helps robots know where they are better. This study
offers several benefits, addressing the challenges of frequent
changes in hospital layouts through mapping and specialized
descriptors. However, it may pose computational demands for
map creation and image quality enhancement.
A notable advancement is the MKR (Muratec Keio Robot),
an advanced robot designed for healthcare [18]. It uses omni-
directional wheels to move safely and efficiently, avoiding
collisions with obstacles. The robot uses virtual potential
fields to navigate globally, considering both stationary and
moving obstacles. Experiments in a hospital demonstrated
its ability to navigate successfully, avoiding collisions. This