To improve the programming efficiency of automatic assembly system, a novel skill programming framework based on task learning is proposed for modular assembly system in this paper. In this framework, the motion sequence of assembly skills can be modeled by demonstration data. And the assembly task is represented hierarchically. A complete assembly process of a part is divided into several skills, and each skill is divided into several sequential assembly motion primitives (AMP) of multiple modules. Then, a learning method of assembly motion sequence based on Hidden Markov Model is proposed, and the maximum probability method is used to generate the optimal sequential AMP. Each AMP is input to the assembly system in the form of instruction to complete the assembly. Aiming at the problem of accurate positioning and trajectory planning, visual guidance and direct teaching method are used to settle this problem. To evaluate the viability of the proposed framework, a customized modular assembly system is used to acquire the demonstration data, and a graphical user interface (GUI) software is designed. Five assembly skills are learned. Experimental are conducted to validate the effectiveness of the proposed method.
This article proposes a robotic precision assembly system for the typical parts of microstructures such as non-silicon microelectromechanical system parts. The assembly system contains three parallel assembly units and support system. In each unit, the images of the base part and target part can be obtained simultaneously from the coaxial alignment vision detection module. This article proposes a system calibration method to ensure accuracy of assembly system. Assembly experiments and accuracy validation tests are conducted. The experimental results show that the synthesizing assembly accuracy of the robotic precision assembly system can reach higher than 3 µm and the mean assembly cycle time of each part is less than 20 s, including the time for feeding and unloading parts. The robotic precision assembly system with nondestructive imaging and gripping can save on cost, achieve better quality results in less time, and greatly facilitate the precision assembly reliability and automation of microstructures.
There are only few works in literature that suggest an assembly process optimization method based on manufacturing errors in the precision manufacturing area. A multi-objective assembly process parameters evaluation and optimization method for precision assembly performance of microstructures with manufacturing errors has been proposed in this paper. Based on the model with manufacturing errors, the ABAQUS software is used for simulation and calculation, and the assembly performance evaluation indexes of the microstructures under different assembly process parameters, such as stress value, stress distribution value and pose offset, are obtained. The mapping model of the key assembly process parameters and assembly performance is established based on BP neural network. Finally, the best assembly process parameters for the optimal assembly performance are solved based on the genetic algorithm, and the method has been verified by the optimization results of preload forces of the 3D mechanism, which can be used to guide and monitor the assembly process quantitatively in the precision manufacturing area.
Aiming at the problems of low assembly knowledge shareability and reusability as well as long generation cycle of assembly process, this paper proposes an ontology-based assembly knowledge representation method, and generates assembly process file based on this method. The assembly ontology, modelling through protégé software, has three central classes: AssemblyObject, AssemblyElement, and AssemblyTool. The assembly ontology is described in OWL language and the assembly knowledge concepts including classes and individuals are linked through properties. In addition, the assembly ontology in OWL language is parsed through Python's RDFLib library, and it is called and displayed in LabVIEW. Finally, the assembly process file containing assembly sequence and assembly process parameters is generated. This method realizes the formal description of assembly process knowledge at the semantic level and improves the shareability and reusability of assembly knowledge. Besides, the corresponding assembly process knowledge can be quickly queried and obtained through this method, improving the efficiency of assembly process planning, and providing intelligent assembly basic knowledge.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.