Details

Autor(en) / Beteiligte

• Fazeli, S. Mahdi
• Abedi, Mostafa
• Molaei, Amir
• Hassani, Masoud
• Khosravi, Mohammad A.
• Ameri, Adel

Titel

Active fault-tolerant control of cable-driven parallel robots

Ist Teil von

• Nonlinear dynamics, 2023-04, Vol.111 (7), p.6335-6347

Ort / Verlag

Dordrecht: Springer Netherlands

Erscheinungsjahr

2023

Quelle

SpringerLink

Beschreibungen/Notizen

• Cable-driven parallel robots (CDPRs) are a class of parallel robots where the cables are used as the arms of the robot in a redundant kinematic architecture. Such an architecture allows manipulation in a considerably large working area and also operation at high speed and acceleration. However, as cables only support tensile force, the controller should generate positive control effort, which leads to a complex constrained control problem. The positive tension distribution in redundant CDPR is generally maintained via redundancy resolution (RR) methods, which are originally optimization-based. The RR methods are prone to kinematic uncertainty, which brings complexity to the control apart from the original effects of kinematic uncertainty. The constrained control of CDPRs gets more complex in the faulty mode of the actuators. The combined effects of faulty actuators along with the kinematic uncertainty limit the application of RR in practice. To address this, a robust fault-tolerant constrained control scheme is proposed to generate an unidirectional control signal for maintaining positive tension in cables in the presence of actuator fault and model uncertainties, taking advantage of an adaptive finite-time sliding mode control along with a nonlinear adaptive observer. The proposed observer estimates the model uncertainties and the actuator fault, while the controller ensures the convergence of the state variables and compensates for the observer estimation error. The H ∞ asymptotic stability of the proposed observer is ensured through sufficient conditions using linear matrix inequality. Furthermore, Lyapunov’s second method is employed to prove the finite-time stability of the system. The performance of the proposed scheme is experimentally validated using a planar CDPR in the presence of actuator fault and model uncertainties.