A Dynamic Obstacle Avoidance algorithm for unmanned aerial vehicles based on Predictive Velocity Obstacles

In increasingly complex UAV operational environments, navigation technology is critical for flight safety. While velocity obstacle methods effectively detect collision risks for uniform linear motion, it is inadequate for real-world nonlinear obstacle movements. To address this limitation, we propos...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Robotics and autonomous systems Jg. 196; S. 105250
Hauptverfasser: Liang, Yanxia, Wang, Huan, Liu, Xin, Zhao, Xiaofan, Ye, Yinghui, Zheng, Yawen, Jiang, Jing
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 01.02.2026
Schlagworte:
Adaptive computational time step
Dynamic Obstacle Avoidance algorithm
Fuzzy controller
Unmanned aerial vehicles
Velocity obstacle method
Unmanned aerial vehicles
Fuzzy controller
Dynamic Obstacle Avoidance algorithm
Velocity obstacle method
Adaptive computational time step
ISSN:0921-8890
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:In increasingly complex UAV operational environments, navigation technology is critical for flight safety. While velocity obstacle methods effectively detect collision risks for uniform linear motion, it is inadequate for real-world nonlinear obstacle movements. To address this limitation, we propose a Dynamic Obstacle Avoidance algorithm for UAVs based on the Predictive Velocity Obstacle method (DOA-PVO), which is applied to a static trajectory. The algorithm expands the velocity obstacle region by merging velocity obstacle circles centered at current and predicted obstacle positions. We introduce an obstacle motion-sensitive evaluation function that selects a more anticipatory velocity by adding penalty costs to candidate velocities in the obstacle’s direction. Furthermore, an adaptive time step is determined using a fuzzy controller, serving both as the interval for calculating new velocities and the duration of flight at the previous velocity. At each new position, an updated expanded velocity obstacle region is computed to assess whether the UAV should return to the static trajectory. Simulation results show the proposed algorithm reduces maneuvering time by up to 55.91%, cumulative angular change by 33.28%, and maneuver count by 57.14%, improving efficiency and trajectory safety. Its robustness is verified via 200-scenario Monte Carlo tests.
ISSN:0921-8890
DOI:10.1016/j.robot.2025.105250