Computer Aided Optimal Robotic Assembly Sequence Generation

Abstract

Robots are widely used for assembly operations across manufacturing industries to attain high productivity through automation. An appropriate robotic assembly sequence further minimizes the total production lead time and overall cost by minimizing the number of assembly direction changes, assembly gripper changes and assembly energy thus selection of a valid optimal robotic assembly sequence is significantly essential to achieve economized manufacturing process. An optimal assembly sequence must comply with various assembly requirements in order to make sure that the sequence of assembly operations is functionally feasible in physical environment.
In order to test an assembly sequence for its practical possibility, necessary assembly information must be collected accurately from the product. Obtaining such assembly information from product drawings or Computer Aided Design (CAD) models in manual mode were involved in lots of complexity and needs high level skills to ensure correctness. Though retrieving such information from products with less number of parts is simple and less time consuming, for products composed of huge number parts it is very complicated and time consuming. Besides retrieving the assembly information, using it for validating an assembly sequence further raises the complexity of the Assembly Sequence Generation (ASG) problem. To perform optimal feasible assembly sequence generation efficiently, an effective computer aided automated method is developed and executed at two phases. The first phase of research is mainly focused on representing the assembly information in a streamlined manner by considering all possible states of assembly configurations for ease of computerization and developing efficient methods to extract the assembly information automatically from CAD environment though Computer Aided Automation (CAA). These methods basically use assembly contact analysis, part transformations and laws of equilibrium & balancing of rigid bodies. From the existing ASG methods, it is observed most of the researchers ignored/not-considered few of the assembly information such as assembly stability data and mechanical feasibility data due to higher complexity in retrieving it from CAD environment.....