Page 253 - 2024-Vol20-Issue2

249 | Sabeeh & Al-Furati

Fig. 1. Basic concept of Kalman filtering.

algorithm is evaluated using three metrics: average
path length, average computational time, and average
smoothness. These metrics offer insights into the effi-
ciency and effectiveness of the algorithm across various
scenarios.

B. Kalman Filtering Fig. 2. Flowchart describing the systematic execution of the
The proposed algorithm, GA-PRM, effectively addresses the GA-PRM algorithm.
issue of balancing collision avoidance and smoothness in path-
planning, especially in dynamic environments like healthcare genetic algorithm (GA) can be overly cautious or too risky,
settings, through the implementation of the Kalman filter [38] affecting navigation speed. To address this, the GA-PRM
for path smoothness. Kalman Filtering is a widely used tech-
nique in the GA-PRM algorithm to improve path planning
accuracy. It is a systematic approach, which continuously up-
dates and refines its understanding of a system’s performance
using sensor data and their uncertainties. This method is
valuable because it combines noisy sensor data with dynamic
models to provide a more reliable assessment of the system’s
status over time. Kalman Filtering operates in two stages: pre-
diction and adjustment based on new measurements, making
it particularly useful for real-time applications.

Figure 1 illustrates the core stages of Kalman Filtering,
encompassing prediction and update. It visually represents
how the filter not only monitors the average state value but
also estimates the degree of variation. As shown in this figure,
the Kalman filter maintains information about the system’s
estimated state and the level of uncertainty associated with
this estimate. This estimation is refined through the utilization
of a model depicting how the state changes over time and the
inclusion of measurements. Specifically, denotes the estimate
of the system’s state at a given time step k before considering
the k-th measurement yk, while represents the corresponding
level of uncertainty.

C. Algorithm Description
In terms of its contribution to the operation of the GA-PRM
algorithm, the Kalman filter is crucial in balancing a robot’s
path to avoid obstacles while maintaining smooth motion. The