Incremental approach for trajectory generation of spray painting robot

Abstract

Purpose This paper aims to present a new offline robot programming approach for automated trajectory generation on free-form surfaces targeted toward spray painting application. Design/methodology/approach In this paper, an incremental trajectory generation approach is developed where new paint passes are generated based on paint deposited on the surface as a result of previous paint passes. The trajectory is generated on real surfaces where optimal velocity is calculated using genetic algorithm considering parameters such as surface model, spray gun model, paint distribution model and task constraints. Findings The developed approach was implemented on various surfaces for different paint distribution patterns, and the simulation results reveal that the approach is flexible and efficient to handle variety in part geometry and paint distribution. From experimental validation and analysis of results thus obtained, the developed approach is highly promising compared to the existing methods. Research limitations/implications The approach assumes that the computer-aided design (CAD) model of the surface is available and is limited to surjective surfaces in a structured environment where the spray gun characteristics and process parameters are known beforehand. Practical implications The problem of programming a robot manually is overcome by automatically generating a sub-optimal trajectory which can be easily transferred to an industrial robot for spray painting the surface. Originality/value This paper discusses a new approach for automated trajectory generation from CAD model. The experimental validation of the developed approach is successfully performed on a highly curved test surface, and obtained results are in agreement with the simulation results.

A Novel Mesh Following Technique Based on a Non-Approximant Surface Reconstruction for Industrial Robotic Path Generation

“…6. Then, vectors a and b can be obtained as the differences Q − P and R − P, respectively, described in Expressions (11) and (12). Thus, the normal vector is calculated with respect to the surface η, as the cross product.…”

“…Results show a maximum position error of 2.7 mm, with a linear velocity of 300 mm/s. MFTs are also proposed for applications such as NDE and painting, using a triangulated surface approximation [11], [12], wherein an incremental trajectory is generated by dividing the surface into different sections, with surface section points and surface-section normals requiring a change in the origin of the part model and the calculation of centroids for every facet. The chosen polygons are sorted to compare their areas with the total area covered by a specific neighboring radius of the traversed trajectories.…”

Fabrication of High-Quality Polymer Composite Frame by a New Method of Fiber Winding Process

“…The optimization of robot trajectory in the fiber winding process is also addressed, in which graph theory is used to obtain the optimal robot trajectory as well as special algorithm such as genetic algorithm, harmony search and also Bézier curves [28][29][30][31][32]. A similar topic, however, focused rather on the study of trajectory control of an arbitrary shape winding mandrel in 3D circular braiding is explored in [20,23,27]. Sofi et al presented dry fiber winding possibilities and explanation of the most important processes of winding [22], and Polini et al pointed out that the tension of winding during robotic filament winding technology is a very important parameter that influences directly the defects and the mechanical property of composite [23].…”

“…Along with specific selections of layers sequence and angle that is made through design process, the "correct winding angle process" is to ensure the arrangement of the fibers such that the angle of the fibers to the frame remains uniform, homogeneous, and consistent in circular-helix cross-sectional form throughout the complex 3D curved geometry of the frame. Determination of the correct trajectory of a robot during winding of the fibers is presented, for example, in [19,[25][26][27]. Gao et al proposed high-speed fiber placement technology in a new methodology of motion planning in a redundant robotic system [25], in which the problem of time optimization of robot motion is solved [25,28,29].…”

“…The optimization of robot trajectory in the fiber winding process is also addressed, in which graph theory is used to obtain the optimal robot trajectory as well as special algorithm such as genetic algorithm, harmony search and also Bézier curves [28][29][30][31][32]. A similar topic, however, focused rather on the study of trajectory control of an arbitrary shape winding mandrel in 3D circular braiding is explored in [20,23,27].…”