A New Task-Space Neural Nonlinear Control Approach for Robotic Manipulators Under Joint Constraints
Robotic manipulators are nonlinear systems with multi-input multi-output (MIMO) structures, uncertainties, and time-varying dynamical prospects. Unexpected influences of internal and external disturbances along with physical constraints in complicated kinematic configurations are large barriers for...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11007530/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Robotic manipulators are nonlinear systems with multi-input multi-output (MIMO) structures, uncertainties, and time-varying dynamical prospects. Unexpected influences of internal and external disturbances along with physical constraints in complicated kinematic configurations are large barriers for deriving excellent controllers of the manipulators. In this paper, we propose a novel two-level control approach to drive the end-effector of robotic manipulators following referenced profiles within the hard bounds of joint angles without using inverse kinematics solutions. A new Cartersian-based high-level sub controller is designed for calculating the desired angular velocities from the task-space control objective throughout a constrained-to-free transformation operator. An advanced sliding mode control framework is adopted to construct the low-level control layer to realize the indirect joint-control missions given from the upper one within the physical bounds. To effectively suppress the dynamical internal and external disturbances, a new neural network with a flexible nonlinear learning law is integrated in the joint-space sub controller. Stability of the whole system including both Cartersian and joint spaces is thoroughly proven by a Lyapunov function under a special constraint. The designed controller was verified on a two-link planar robotic arm in various challenging conditions and in comparing with other state-of-the-art control methods. The in-depth simulation results exhibited that the proposed control algorithm possesses outperformance working abilities such as model-free adaptation, robustness, high accuracy and comprehensiveness. |
|---|---|
| ISSN: | 2169-3536 |