Adaptive control for robot manipulators using multiple parameter models

Shafiqul Islam, Peter X. Liu, Jorge Dias, Lakmal D. Seneviratne

Research output: Contribution to journalArticlepeer-review

8 Scopus citations


In this paper, we propose multiple parameter models based adaptive switching control system for robot manipulators. We first uniformly distribute the parameter set into a finite number of smaller compact subsets. Then, distributed candidate controllers are designed for each of these smaller compact subsets. Using Lyapunov inequality, a candidate controller is identified from the finite set of distributed candidate controllers that best estimates the plant at each instant of time. The design reduced the observer-controller gains by reducing modeling errors and uncertainties via identifying an appropriate control/model via choosing largest guaranteed decrease in the value of the Lyapunov function energy function. Compared with CE based CAC design, the proposed design requires smaller observer-controller gains to ensure stability and tracking performance in the presence of large-scale modeling errors and disturbance uncertainties. In contrast with existing adaptive method, multiple model based distributed hybrid design can be used to reduce the energy consumption of the industrial robotic manipulator for large scale industrial automation by reducing actuator input energy. Finally, the proposed hybrid adaptive control design is experimentally tested on a 3-DOF PhantomTM robot manipulator to demonstrate the theoretical development for real-time applications.

Original languageBritish English
Pages (from-to)1365-1375
Number of pages11
JournalInternational Journal of Control, Automation and Systems
Issue number5
StatePublished - 1 Oct 2016


  • Adaptive switching control
  • Lyapunov method
  • multiple models
  • observer


Dive into the research topics of 'Adaptive control for robot manipulators using multiple parameter models'. Together they form a unique fingerprint.

Cite this