Trajectory planning of redundant parallel mechanism considering motion accuracy based on reinforcement learning

The motion accuracy of the trajectory directly affects the reliability of parallel mechanisms in precision tasks such as micro-assembly. Thus, this paper investigates trajectory planning considering motion accuracy, and uses a (6+3)-degrees of freedom (DOF) kinematically redundant parallel mechanism...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Engineering applications of artificial intelligence Jg. 162; S. 112552
Hauptverfasser: Zeng, Chen-dong, Qiu, Zhi-cheng, Zhang, Fen-hua, Zhang, Xian-min
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier Ltd 22.12.2025
Schlagworte:
Motion accuracy
Redundant parallel mechanism
Reinforcement learning
Trajectory planning
Twin delayed deep deterministic policy gradient algorithm
Motion accuracy
Trajectory planning
Redundant parallel mechanism
Reinforcement learning
Twin delayed deep deterministic policy gradient algorithm
ISSN:0952-1976
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The motion accuracy of the trajectory directly affects the reliability of parallel mechanisms in precision tasks such as micro-assembly. Thus, this paper investigates trajectory planning considering motion accuracy, and uses a (6+3)-degrees of freedom (DOF) kinematically redundant parallel mechanism (KRPM) as a case study. First, the kinematics of KRPM are analyzed, and an error model incorporating dimensional errors, driving input errors, and joint clearances is established. Then, based on the error model, the aggregate sensitivity index (ASI) and comprehensive error sensitivity (CES) are introduced to study the error properties of KRPM, along with a universal analysis process. Subsequently, reinforcement learning (RL) utilizing the twin delayed deep deterministic policy gradient (TD3) algorithm is employed to the trajectory planning of KRPM considering motion accuracy. Finally, numerical simulation is carried out based on three cases to verify the effectiveness of the proposed method, and experimental results further demonstrate its practical applicability. •The error model of a (6+3)-DOF kinematically redundant parallel mechanism is established.•The trajectory planning problem considering motion accuracy is modeled as MDP and solved by the TD3 algorithm.•Numerical simulations are carried out based on three cases to verify the effectiveness of the proposed method.•Experimental results further verify the practical applicability of the proposed method.
ISSN:0952-1976
DOI:10.1016/j.engappai.2025.112552