| NSF Org: |
CMMI Division of Civil, Mechanical, and Manufacturing Innovation |
| Recipient: |
|
| Initial Amendment Date: | August 1, 2014 |
| Latest Amendment Date: | August 1, 2014 |
| Award Number: | 1426799 |
| Award Instrument: | Standard Grant |
| Program Manager: |
Bruce Kramer
CMMI Division of Civil, Mechanical, and Manufacturing Innovation ENG Directorate for Engineering |
| Start Date: | September 1, 2014 |
| End Date: | August 31, 2018 (Estimated) |
| Total Intended Award Amount: | $300,000.00 |
| Total Awarded Amount to Date: | $300,000.00 |
| Funds Obligated to Date: |
|
| History of Investigator: |
|
| Recipient Sponsored Research Office: |
77 MASSACHUSETTS AVE CAMBRIDGE MA US 02139-4301 (617)253-1000 |
| Sponsor Congressional District: |
|
| Primary Place of Performance: |
77 Massachusetts Ave Cambridge MA US 02139-4307 |
| Primary Place of
Performance Congressional District: |
|
| Unique Entity Identifier (UEI): |
|
| Parent UEI: |
|
| NSF Program(s): | NRI-National Robotics Initiati |
| Primary Program Source: |
|
| Program Reference Code(s): |
|
| Program Element Code(s): |
|
| Award Agency Code: | 4900 |
| Fund Agency Code: | 4900 |
| Assistance Listing Number(s): | 47.041 |
ABSTRACT
Robots for application in collaborative manufacturing must perform manual work side-by-side with people. Such robots offer the flexibility to work on many different tasks and promise to transform manufacturing by improving the quality and efficiency of manual processes in small shops and in facilitates that manufacture highly customized products. However, in order to meet this promise, robots must be effectively integrated into existing manufacturing teams and practices. To enable this integration, this National Robotics Initiative (NRI) award supports fundamental research on the methods and instruments that manufacturing engineers will need to form effective human-robot teams based on task requirements and worker skills. These methods will also enable robots to adapt to changes in workflow to maximize safety and efficiency. The effective integration of collaborative robots into manufacturing promises improvements in many industries that have not yet benefited from robotic technology. Therefore, results from this research will contribute to the competitiveness of U.S. manufacturing and benefit the U.S. economy and society. The research will involve contributions from multiple disciplines, including robotics, human factors, computer science, and manufacturing, and by academic and industry collaborators. These collaborations will help the dissemination of research results into manufacturing organizations and the integration of research into undergraduate and graduate curriculum in engineering.
Advancements in robotics promise the use of collaborative robots that perform interdependent work with people in order to improve quality, efficiency, and safety in industrial manufacturing. However, integrating collaborative robots into these processes and ensuring their efficient operation pose significant research challenges, including the optimal allocation of work based on task requirements and constraints, the formation of human-robot teams, and the dynamic adaptation of teamwork to workflow changes. This research will address these research challenges, enabling the seamless integration of collaborative robots into these processes and achieving efficient and safe collaboration between human and robot workers. The research team will create novel methods for optimal allocation of tasks to human and robot workers based on task constraints and worker skills, design new tools that utilize these methods to facilitate workflow design for human-robot teams, and develop novel mechanisms that enable robots to more efficiently and safely collaborate with human workers in the planned manufacturing operations. These methods and instruments will be validated in real-world manufacturing operations and disseminated through industry workshops, engineering curricula, and a public outreach program.
PUBLICATIONS PRODUCED AS A RESULT OF THIS RESEARCH
Note:
When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external
site maintained by the publisher. Some full text articles may not yet be available without a
charge during the embargo (administrative interval).
Some links on this page may take you to non-federal websites. Their policies may differ from
this site.
PROJECT OUTCOMES REPORT
Disclaimer
This Project Outcomes Report for the General Public is displayed verbatim as submitted by the Principal Investigator (PI) for this award. Any opinions, findings, and conclusions or recommendations expressed in this Report are those of the PI and do not necessarily reflect the views of the National Science Foundation; NSF has not approved or endorsed its content.
The major goals of the project are to design, develop, and evaluate methods and tools that enable the seamless integration of collaborative robots into human-robot teams and achieve efficient and safe collaboration between human and robot workers in industrial manufacturing. To meet these goals, the project team undertook the following major research activities:
1. Thread 1 (Integration Methods): Creating novel methods for more optimally allocate tasks to human and robot workers based on task constraints and worker skills,
2. Thread 2 (Integration Tools): Designing new tools that utilize these methods to facilitate workflow design for human-robot teams.
3. Thread 3 (Coordination Mechanisms): Developing novel mechanisms that enable robots to more efficiently and safely collaborate with human workers in the planned manufacturing operations
To advance integration methods, we developed a method for multi-objective human-robot collaborative task planning. We also devised a real-time system to provide human workers with performance feedback and/or performance-based updated task plans.
Our exploration of methods for multi-objective task planning showed that an off-the-shelf planner can quickly return a ?satisficing? solution, which is important for replanning when the work environment and conditions change in unexpected ways. Additionally, we developed a visualization tool that communicates the generated plan to the user in the form of a Gantt chart, and integrated a method of estimating worker ergonomics in real-time.
To advance human in-the-loop planning, we conducted a laboratory study involving 36 participants that analyzed human performance on task allocation and scheduling problems. We tested benchmark problems defined from the International Planning Competition and compared plans generated by users against plans generated by a state-of-the-art planner. In the study, both planners were given 30 minutes to come up with a best plan. We found that human-generated plans, on average, outperformed computer-generated plans, up to 40% in plan quality. Our results showed that users excelled in domains where the problem can be visualized onto a map (e.g., vehicle
routing). We also found that users were quick at detecting the most constraining resource and then optimizing plans around that resource.
The finding that humans can outperform state-of-the-art planners strongly support a collaborative framework.
To advance integration tools and coordinationg mechanisms, we developed a system for processing a structured form of human team's planning conversation to infer the discussed plan, and an authoring environment for adapting the plan for human-robot collaborative work.
Last Modified: 02/26/2019
Modified by: Julie Shah
Please report errors in award information by writing to: awardsearch@nsf.gov.
