Base position optimization for mobile painting robot manipulators with multiple constraints. (December 2018)

• Record Type: Journal Article
• Authors: Yu, Qiankun; Wang, Guolei; Hua, Xiaotong; Zhang, Simin; Song, Libin; Zhang, Jiwen; Chen, Ken
• Is Part Of: Robotics and computer-integrated manufacturing. Volume 54(2018)
• Page Start: 56
• Page End: 64
• Publication Date: 2018-12
• Subjects: Base position optimization -- Mobile robot manipulator -- Painting robot -- Approximate decoupled model -- Inequality constrained optimization problem; Robots, Industrial -- Periodicals; Computer integrated manufacturing systems -- Periodicals; Robotics -- Periodicals; Robots industriels -- Périodiques; Productique -- Périodiques; Robotique -- Périodiques; 670.285
• DOI: 10.1016/j.rcim.2018.05.007
• Languages: English
• ISSNs: 0736-5845

Abstract

Highlights: Propose an efficient base position optimization methodology for the task planning of mobile robot manipulators painting large workpieces. Establish an approximate decoupled model for painting robot manipulators with high accuracy. Constraints summarized from manipulating tasks and systems guarantee the result available. The internal penalty function method and the generalized Lagrange multiplier method are modified to get a near-optimal feasible BP. Application results indicate that the efficiency is improved by hundreds of times in comparison with previous method. Abstract: This paper proposes an efficient base position (BP) optimization method for mobile painting robot manipulators (MPRMs). An approximate decoupled model is first established to overcome the coupling problem of painting robots. And the manipulating characteristics are summarized as three constraints: positioning, orientation and singularity avoidance constraints. Then, joint-level performance criteria of one manipulating point and a painting path, which reflect the manipulability and dexterity, were constructed successively. Considering multiple constraints, the BP optimization problem is translated into a standard inequality constrained optimization problem of the path criterion. Two algorithms are designed to solve this problem: one is based on the internal penalty function method used to obtain an initial BP; the other is based on the generalized Lagrange multiplier method used to get the near-optimal BP. This method was applied to a real MPRM system painting three typical surfaces: flat, cylindrical and truncated conical surfaces. Application results demonstrate the effectiveness as well as the availability of the approximate decoupled model. Simultaneously, compared with previous methods, the efficiency is improved by hundreds of times.