A Diffusion-Augmented GWO-TCN-PSA Method for Real-Time Inverse Kinematics in Robotic Manipulator Applications
Year: 2026
Type: article
Abstract: This paper presents an efficient inverse kinematics (IK) solution for robotic manipulators, addressing the challenges of high computational complexity, low efficiency, and sensitivity to singularities associated with traditional methods. A data augmentation strategy is introduced, utilizing an enhanced Diffusion-TS model to generate diverse joint-angle samples and corresponding end-effector poses through forward kinematics, thereby creating a high-quality dataset. To improve real-time performance, a Temporal Convolutional Network (TCN) model is developed, optimized using the Grey Wolf Optimizer (GWO), and augmented with a probabilistic sparse attention mechanism to effectively capture key pose features. Experimental evaluations on the Jaka MiniCobo robotic arm demonstrate that the proposed method significantly reduces inference time while maintaining high accuracy, making it suitable for real-world applications that demand both speed and precision.
Source: Electronics
Institution: Shandong University
Language: English
Subfield: Control and Systems Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W7154596176