Advances in robotics and sensing have the potential to dramatically increase the productivity of US manufacturing. A barrier to achieving this potential is the high cost to integrate the components of a robot workcell, typically an order of magnitude higher than the cost of the robot and equipment. The added cost arises from the difficulty in integrating the robot controller with the controllers of the auxiliary equipment in the workcell, and with neighboring workcells in the factory. This project will develop measurement science for control architectures, sensor network protocols, and interface standards that reduce the time and cost to integrate robots with other factory automation systems, and robots and automation systems with factory-level control systems.
Reduce the time, effort, and expense needed to configure and implement manufacturing robotic workcells by developing and deploying measurement science for workcell control architectures and protocols. Deliver results through standards bodies by 2014.What is the new technical idea?
The new technical idea is to develop the measurement science needed to evaluate the effectiveness of architectures, languages, and protocols that enable robot workcells to be designed, configured and programmed quickly and easily using commercial off the shelf components that conform to standard architectures and protocols. This vision is challenged by systems that take too long to reconfigure due to lack of plug and play compatibility; robot programming that is slow and sensitive to small changes arising from proprietary languages and architectures; and the inability to re-use existing automation components in reconfigured workcells to meet new process requirements. This project will address these challenges by working with robot system end-users and vendors to develop and deploy standard protocols, languages and interfaces. The project will establish a robot integration testbed for measuring the effectiveness of standards for the rapid reconfiguration of robots and auxiliary equipment in real, virtual and hybrid modes.What is the research plan?
The project work will take place in three phases: problem analysis and testbed development, test development, and standardization. In the analysis phase the state of the industry practice and needs for robot workcell integration will be analyzed and classified according to best methods of improvement, including standard information models, standard interfaces and operating systems, and use of simulation and emulation tools. This analysis will guide the formulation of a robot integration testbed that supports testing integration of components using new protocols, and conducting experiments in rapid reconfiguration of robots and auxiliary equipment in real, virtual and hybrid modes. Research issues include designing its meta-architecture (how it can easily be reconfigured to support different factory architectures), and defining and implementing its computing, networking, modeling and simulation resources. NIST will work with industry associations such as Automotive Industry Action Group, the Robotic Industries Association and the Association for Manufacturing Technology to ensure that NIST work focuses on relevant US manufacturer problems.
During the test development phase, the research challenge is defining tests that measure the degree to which different factory architectures, protocols and information exchange methodologies help reduce integration costs. Tests will involve industrial components when available, or component emulations generated by vendors, integrators, or NIST. Testbed infrastructure will be used to integrate components and test their interactions in industry generated scenarios. Validation tests will measure how completely a candidate protocol supports the targeted robotic applications. Performance tests will measure the impact of standards on resource requirements, computing and communication infrastructure. NIST will identify gaps between requirements and actual performance, and report results to partner organizations. Additional test outputs can include new requirements discovered, and retest after improvements/fixes are made.
During the standardization phase this project will work with robot system end-users and vendors to define the tests that measure conformance of standard protocols, languages, and interfaces. The research work will develop the test cases that are the basis for conformance tests that evaluate how correctly products meet the standard. The NIST contribution will help ensure that the protocols, languages, and interfaces adequately cover state-of-the-art functionality (e.g., position and force control, vision-guided motion) for important manufacturing processes (e.g., material handling, assembly, welding and painting). The final outcomes will be standards supported by industrial automation vendors and adopted by manufacturing enterprises.Recent Results:
This project began in the middle of FY 2012; the following goals are in progress:
This project will contribute to MTConnect version 1.3 in the area of Robotics Device Integration, submitting revisions resulting from validation testing to the MTConnect Technical Advisory Group, the Standards Drafting Group, and the Robotics Device Integration Group in which project staff already participate. Formal standardization of MTConnect and ROS-Industrial will be explored through ANSI-accredited organizations such as the Robotic Industries Association (RIA). NIST already participates in AMT and RIA committees, and plans on participating in the ROS-Industrial Consortium.
Start Date:October 1, 2012
Lead Organizational Unit:el
Related Programs and Projects:
Bill Rippey, Project Leader