A Dynamic Density Peak Clustering Algorithm Based on K-Nearest Neighbor

Du, Hui; Zhai, Qiaofeng; Wang, Zhihe; Li, Yongbiao; Zhang, Manjie

doi:10.1155/2022/7378801

Cited by 3 publications

(2 citation statements)

References 22 publications

(18 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This method relies heavily on human observation, which makes it ineffective for the clustering of non-uniform density datasets. A dynamic density peak clustering algorithm based on the K-nearest neighbor (DDPC) method was proposed in [42]; this approach can reduce the parameter sensitivity and choose cluster centers automatically. In [43], the best Eps parameter neighborhood values were selected using the global search capability of the bird swarm method.…”

Section: Introductionmentioning

confidence: 99%

Path Planning Algorithm Based on Obstacle Clustering Analysis and Graph Search

Wang

Sun

2023

Symmetry

View full text Add to dashboard Cite

Path planning is receiving considerable interest in mobile robot research; however, a large number of redundant nodes are typically encountered in the path search process for large-scale maps, resulting in decreased algorithmic efficiency. To address this problem, this paper proposes a graph search path planning algorithm that is based on map preprocessing for creating a weighted graph in the map, thus obtaining a structured search framework. In addition, the reductions in the DBSCAN algorithm were analyzed. Subsequently, the optimal combination of the minPts and Eps required to achieve an efficient and accurate clustering of obstacle communities was determined. The effective edge points were then found by performing obstacle collision detection between special grid nodes. A straight-line connection or A* planning strategy was used between the effective edge points to establish a weighted, undirected graph that contained the start and end points, thereby achieving a structured search framework. This approach reduces the impact of map scale on the time cost of the algorithm and improves the efficiency of path planning. The results of the simulation experiments indicate that the number of nodes to be calculated in the search process of the weighted graph decreases significantly when using the proposed algorithm, thus improving the path planning efficiency. The proposed algorithm offers excellent performance for large-scale maps with few obstacles.

show abstract

Section: Introductionmentioning

confidence: 99%

Path Planning Algorithm Based on Obstacle Clustering Analysis and Graph Search

Wang

Sun

2023

Symmetry

View full text Add to dashboard Cite

show abstract

“…Tis article has been retracted by Hindawi following an investigation undertaken by the publisher [1]. Tis investigation has uncovered evidence of one or more of the following indicators of systematic manipulation of the publication process:…”

mentioning

confidence: 99%