<![CDATA[This paper presents the design and performance testing of the Enhanced Informed Probabilistic Roadmap (EI-PRM) algorithm in path planning in various environments, such as simple environments, dense environments and narrow paths. This research evaluates the effectiveness of the algorithm in terms of solution cost and computation time by testing different parameter configurations, including number of sample points (nsample) and cost scaling factor (). The results show that the EI-PRM algorithm can adjust the sampling strategy based on the available information, resulting in an optimal solution with high efficiency. During the test, in a simple environment with the parameter value of nsample between 200 and 400 and parameter value between 1.2 and 1.4, the best solution cost is 344.93 and the computation time is 1.9 seconds. However, in a denser environment, the optimal solution cost reaches 141,586 with a computation time of 1.16 seconds, a parameter value of nsample 200, and a parameter value of 1.5. Furthermore, the algorithm shows good performance on narrow paths with an optimal solution cost of about 293.39 and the best computation time of 0.38 seconds at a parameter value of nsample 400 parameter 1.3. This research focuses on the importance of parameter optimization and efficient sampling strategies to improve path quality and speed up computation time. In general, the results indicate that the EI-PRM algorithm is effective for path planning under various environmental conditions. The process of the EI-PRM algorithm consists of several steps. First, sample points are created at random. In the second step, the computer will link the example locations to produce a roadmap. In the last step, the shortest path inside an ellipsoid-bounded search area will be determined. The size of the ellipsoid will increase gradually until the best path solution is found. This research is expected to contribute significantly to the development of path planning algorithms that are more efficient, faster and capable of producing high-quality paths in complex environments. This research has the potential to improve applications in transportation and logistics that require optimal path planning in order to reduce operational costs and improve safety.]]>
Copyrights © 2025
