A software platform to develop and execute kitting tasks on industrial cyber-physical systems

Research output: ResearchPh.D. thesis

Abstract

The current material handling infrastructure associated with manufacturing and assembly operations still register a great presence of human work for highly repetitive tasks. A major contributing factor for the low automation is that current manufacturing robots have little or no understanding of the world around them and no capability to dynamically change their actions if the environment or task changes. This restricts them to operate in highly constrained environments and makes it difficult and expensive to change the robot operation from one task to another.

In this dissertation we investigate the use of autonomous mobile manipulators to manage the industrial kitting operation, the task of collecting parts from the warehouse into kits for the production line. We discuss in particular the architectural and implementation aspects of a skill-based control platform that enables the mobile manipulators to accomplish their task autonomously and with a dynamic adaptation to the context.

We show how the platform supports skill programming and configuration, maintains a world model with all relevant information to make autonomous task planning, interfaces with the Manufacturing Execution System (MES) to receive kitting orders and finally executes and monitors the skills to accomplish the orders.

The platform is the foundation and guideline of a concept that in the future could completely revolutionize the robot industries, similarly to how iOS and android revolutionized the industry of smart phones. A platform where skills, similarly to computer or smartphone applications, can be installed and removed from heterogeneous robots with few elementary steps.
Close

Details

The current material handling infrastructure associated with manufacturing and assembly operations still register a great presence of human work for highly repetitive tasks. A major contributing factor for the low automation is that current manufacturing robots have little or no understanding of the world around them and no capability to dynamically change their actions if the environment or task changes. This restricts them to operate in highly constrained environments and makes it difficult and expensive to change the robot operation from one task to another.

In this dissertation we investigate the use of autonomous mobile manipulators to manage the industrial kitting operation, the task of collecting parts from the warehouse into kits for the production line. We discuss in particular the architectural and implementation aspects of a skill-based control platform that enables the mobile manipulators to accomplish their task autonomously and with a dynamic adaptation to the context.

We show how the platform supports skill programming and configuration, maintains a world model with all relevant information to make autonomous task planning, interfaces with the Manufacturing Execution System (MES) to receive kitting orders and finally executes and monitors the skills to accomplish the orders.

The platform is the foundation and guideline of a concept that in the future could completely revolutionize the robot industries, similarly to how iOS and android revolutionized the industry of smart phones. A platform where skills, similarly to computer or smartphone applications, can be installed and removed from heterogeneous robots with few elementary steps.
Original languageEnglish
Number of pages158
StatePublished - 2017
Publication categoryResearch

Bibliographical note

No e-publishing at this time.

    Research areas

  • task planning, knowledge representation and reasoning, robot control architecture, skill-based programming, kitting, industrial mobile manipulators
ID: 260312464