A non-linear model predictive controller with obstacle avoidance for a space robot

Mingming Wang, Jianjun Luo, Ulrich Walter

Research output: Contribution to journalArticlepeer-review

107 Scopus citations

Abstract

This study investigates the use of the non-linear model predictive control (NMPC) strategy for a kinematically redundant space robot to approach an un-cooperative target in complex space environment. Collision avoidance, traditionally treated as a high level planning problem, can be effectively translated into control constraints as part of the NMPC. The objective of this paper is to evaluate the performance of the predictive controller in a constrained workspace and to investigate the feasibility of imposing additional constraints into the NMPC. In this paper, we reformulated the issue of the space robot motion control by using NMPC with predefined objectives under input, output and obstacle constraints over a receding horizon. An on-line quadratic programming (QP) procedure is employed to obtain the constrained optimal control decisions in real-time. This study has been implemented for a 7 degree-of-freedom (DOF) kinematically redundant manipulator mounted on a 6 DOF free-floating spacecraft via simulation studies. Real-time trajectory tracking and collision avoidance particularly demonstrate the effectiveness and potential of the proposed NMPC strategy for the space robot.

Original languageEnglish
Pages (from-to)1737-1746
Number of pages10
JournalAdvances in Space Research
Volume57
Issue number8
DOIs
StatePublished - 15 Apr 2016

Keywords

  • Non-linear model predictive control
  • Obstacle avoidance
  • Space robot

Fingerprint

Dive into the research topics of 'A non-linear model predictive controller with obstacle avoidance for a space robot'. Together they form a unique fingerprint.

Cite this