*
Bookmark and Share

Next-Generation Robotics and Automation Program

Summary:

This program addresses the measurement science and standards necessary to enable the next generation of smart robots and automation systems and to ensure their adoption in manufacturing in the U.S. at scales ranging from the very small to the very large. The program addresses major barriers including perception, manipulation, human-robot interaction, and safety. It does so through targeted projects that develop standards and performance measures that will help drive research and spur development and applications of intelligent automation that can operate in real world manufacturing and in close proximity with people. The targeted manufacturers include both large manufacturers (automotive and aerospace) and the small and medium-sized enterprises that employ 60% of manufacturing workers.

Description:

Objective:

Develop and deploy advances in measurement science to safely increase the versatility, autonomy, and rapid re-tasking of intelligent robots and automation technologies for smart manufacturing and cyber-physical systems applications.

What is the problem?

The U.S. needs to retain its leadership position in manufacturing, which is being whittled away by cheap labor and growing sophistication of foreign competitors[1]. Traditionally, U.S. manufacturers have been seen as innovators but they have not always been able to make products as cost-effectively as is required in the global economy. In order to remain competitive, new ways must be found to make products faster, better, and cheaper. Perhaps the most promising approach comes from the new generation of robotics and automation. This has been recognized to a much greater extent outside the U.S. than within[2]. The European Commission spent €536 M (~$750 M) in the period 2007-2012 for cognitive and robotics related research[3], much of it for manufacturing. Japan spends similar amounts, although until recently their focus was on humanoid robots for service applications. They are now refocusing on manufacturing.

According to a major research roadmap published by the Computing Community Consortium (CCC) based on a study funded by NSF[4], robotics "clearly represents one of the few technologies capable in the near term of building new companies and creating new jobs." The roadmap report concludes that "robotics technology holds the potential to transform the future of the country and is likely to become as ubiquitous over the next few decades as computing technology today." The Office of Science and Technology Policy (OSTP) and Office of Management and Budget (OMB) have called for focused support of R&D in "advanced manufacturing to strengthen U.S. leadership in the areas of robotics, cyber-physical systems, and flexible manufacturing" as a means to promote sustainable economic growth and job creation[5]. The President's Framework for Revitalizing American Manufacturing recognizes the importance of "developing advanced robotics technologies that allow the U.S. to retain manufacturing and respond rapidly to new products and changes in consumer demand"[6].

Robotics technology today consists of a patchwork of ad hoc solutions devoid of a measurement science based framework. The potential of robotics as a transformative technology cannot be attained without a concerted scientific program that includes a robust measurements and standards infrastructure. The participating experts in the CCC Roadmap effort noted that certain essential enabling capabilities are required across the board, including robust 3D perception, planning and navigation, human-like dexterous manipulation, and intuitive and safe human-robot interaction, especially working in proximity to or in direct contact with people. A recent analysis distilled the technology development recommendations from four robotics roadmaps, including the one by CCC, into a single document[7]. The aforementioned list of cross-cutting enabling capabilities was common across all roadmaps. NIST research and development in this program provides the missing performance measures and standards needed in these key technology areas in order to catalyze and accelerate progress toward development and deployment of next generation robotics and automation systems. Several leading thinkers, including Bill Gates, believe that we are on the verge of an era where robotic devices will become ubiquitous in much the same way that PC's have.[8] This is due to the increasing integration of sensors and other advancements that enable robots to function in and adapt to unconstrained and varying environments and in close proximity to humans. These dawning capabilities raise the urgency of preparing a solid foundation with which to quantitatively measure the performance of robotic and automation systems.

The robotics industry started in the United States, but as a result of sustained commitment by others, 80% of the present industrial market is served by European and Asian companies[9]. Now that robotics technology is poised to transform other industries, such as nano-manufacturing and construction, the stakes are much higher. Advances in robotics stand to impact nearly every aspect of U.S. industry and everyday life. Robotics itself is a growth industry as well as an enabler of new industries (e.g., surgical robots, environmental monitoring micro-robots). Traditional manufacturers will benefit from improved quality, greater agility, accelerated times to market, and retention of intellectual property within the U.S. Robotics will improve existing medical procedures and enable new ones that are less invasive and produce fewer side effects. Service robotics – both professional (e.g., agricultural, fire-fighting, construction, etc.) and personal (e.g., domestic, assistive) – in particular will benefit broad segments of the population. Professional service robotics is expected to serve as a productivity multiplier for increased economic growth, while domestic service robotics is expected to enable sustained quality of life through personal independence with age. This effort is critical for the U.S. technological and economic well being: according to the CCC study, "the rest of the world has recognized the irrefutable need to advance robotics technology and have made research investment commitments totaling over $1 billion; the U.S. investment in robotics technology, outside unmanned systems for defense purposes, remains practically non-existing."

Why is it hard to solve?

Addressing this problem is difficult because the multi-disciplinary nature of robots, coupled with the diversity of application requirements, results in a fragmented industry, lacking cohesion and agreement on technological approaches, requirements, and solutions. In each target user community, measurement science is lacking for specifying quantitatively what performance is needed from robotic and automation systems and there are no means of objectively and reproducibly measuring how well robots and automation systems meet the requirements.[10] The robotics market has traditionally been heavily focused on automotive applications. With the expansion into other areas, such as food processing, aerospace, and packaging, there are now a wider set of problems to address and a larger number of relatively inexperienced industry end-users to serve.

How is it solved today, and by whom?

Each of the disciplines required for robotics and automation systems has developed largely independently, mostly focused on non-manufacturing applications, and few organizations have people with skills in all the necessary technologies. This has led to a fragmentation of the industry into three tiers: manufacturers, users, and integrators. Robot and sensor manufacturers develop automation products. Users buy the products but typically lack the skill to install them in their factories and make them productive. Integrators usually custom-build the robotic work cells, retaining the expertise after the products are operational and frequently continuing to manage the applications over their entire lifetimes. This is very expensive and is only cost-effective for large manufacturers who will use a particular work cell for a considerable length of time. Small and medium sized manufacturers, and others that require rapid changeovers, have typically not adopted robotic solutions for this reason.

It is therefore not surprising that robots have penetrated only a small segment of the manufacturing domain. It is estimated that 45% of the robot supply is taken up by 10% of the industry, primarily by companies that have more than 500 employees (mostly automotive), and that 90% of the potential users have not adopted robotics for manufacturing[11]. Further penetration within manufacturing is limited by cost-prohibitive infrastructure required for robot installation. Since 90% of industrial robots today lack any ability to sense their external environment[12], the world in which they function must be perfectly structured to match the pre-programmed instructions that the robots blindly execute. "Industrial robots still do not have the sensing, control and decision making capability that is required to operate in unstructured, 3D environments. Cost-effective, reliable force sensing for assembly still remains a challenge. Finally we still lack the fundamental theory and algorithms for manipulation in unstructured environments, and industrial robots currently lack dexterity in their end-effectors and hands"[13]. These capability gaps severely limit current use of robots in manufacturing.

Potential users face uncertainty when comparing robot suppliers, implementation plans, and safeguarding mechanisms due to the lack of performance specifications, standard test methods, and technical guidance. Prospective buyers also face a complicated landscape when assessing the potential to integrate robots into their existing assembly lines or to incorporate the use of multiple robots in a single application[14]. Existing manufacturers who wish to purchase robots and associated equipment today have to rely on vendors' claims of performance, capabilities, and suitability for a particular application. There are no standard means of verifying performance claims or equitably comparing candidate solutions. The lack of standards is hindering robot developers as well. Currently, they have to build robots for each new application from scratch, resulting in very expensive development costs. Robotics executives have stated that "Robotics would also benefit from working more closely together to create technology standards "[15]. In some countries there are organizations, such as the Fraunhofer Institutes in Germany and the Industrial Technology Research Institute in Taiwan, that help companies overcome some of these problems, but not in the U.S.

Why NIST?

The program directly supports the Engineering Laboratory's mission to anticipate and meet the measurement science and standards needs of U.S. industries conducting technology-intensive manufacturing. It is aligned with the Laboratory's goal to enable the next generation of innovative and competitive manufacturing, construction, and cyber-physical systems through advances in measurement science, and addresses the core competence "Intelligent sensing, control, processes, and automation for cyber-physical systems".

Progress in robotics and automation systems is currently hindered by the lack of ability to specify quantitatively the performance needed from a robot and objectively and reproducibly measure how well these systems meet the performance requirements.[16],[17] The lack of hardware and software platform standards impedes collaborative progress within the research community, makes it difficult to assess experimental results, and leads to unnecessary duplication of engineering and integration effort.  This program applies NIST's measurement science and standards expertise together with its robotics expertise, to ensure that U.S. industries successfully leverage the huge potential of robotics and automation technologies. NIST is uniquely positioned to foster collaboration between users (including government agencies), robot manufacturers, academics, and other researchers to identify common robotics and automation measurement science needs, develop standard test methods to address them, and encourage end-user adoption in factory applications. NIST is actively involved in safety standards for robots in several domains[18], so has the necessary contacts and exposure to manufacturers, integrators and users that will allow successful transfer of knowledge to industry.

What is the new technical idea?

The key idea is to develop and apply performance metrics and associated standard test methods to isolate and measure robot performance against specific application requirements, thereby accelerating the flow of innovations that address U. S. manufacturing industry's needs. NIST has pioneered a process model by which end-user requirements are rigorously assembled, converted to validated metrics, and propagated through reproducible and quantifiable standard test methods. Researchers and manufacturers have proven that they can use their ingenuity to solve the user community's needs by implementing innovative solutions to meet the challenges posed by consensus standard test methods. In this approach, progress against functional or performance goals is measured – design choices are left to the creativity and innovation of the developers. This process model has been successfully applied to a DHS-funded project to develop performance standards for robots for urban search and rescue (USAR)[19]. The model of accelerating innovation in robotics and automation systems, linking application requirements to robot developers through performance metrics and standard test methods, can be expanded to critical, high-impact robot manufacturing applications

Why can we succeed now?

The time is right because of the increased recognition that robotics will become a driving force in industry and society and it is crucial for the United States to recapture and retain the innovation leadership in this growing enterprise. It is also the right time because several necessary enabling technologies and tools are becoming mature or cost-effective enough to become incorporated into robotics and automation systems to address the above-mentioned gaps. For instance, computational power is becoming affordable and powerful enough to support the type of processing and decision-making required for robots to become adaptive to their environment. A new generation of sensors that provides three-dimensional information about the robot's surroundings is becoming mature enough to address the requirement that "perception systems for automation in dynamic environments will need to be comprehensive, pervasive, and redundant "[20]. Advances in the defense robotics domain have enabled robots to perceive and navigate within complex and unpredictable environments. The sensors and algorithms from the defense domain can be harvested for industrial and other applications, but they are not directly transferable.

What is the research plan?

NIST will work closely with the manufacturing user communities to capture their needs and requirements for robotics and automation systems and coordinate the various organizations that will create the next generation of robotics to meet the needs of manufacturing. NIST knowledge will be leveraged both in application areas and component technologies (e.g., 3D sensors, manipulators, human-robot interaction). This program will elicit detailed performance requirements from manufacturers (large, medium, and small), develop preliminary test methods to measure robot performance against requirements, and iterate with the end-user and technology provider communities to assess and refine the metrics and measurement techniques and implement standards.

Work in the program is organized around four thrusts, each of which may have one or more projects at any one time addressing different aspects of the critical technologies for next generation robotics and automation. Thrusts have longer duration than projects. They serve to ensure continuity as projects addressing different aspects of the thrust are completed. Each thrust and project will involve manufacturing industry end-user engagement in the planning, execution, and delivery phases of the work. The thrusts included in the program are:

  • Sensing and Perception for Manufacturing Applications: Sensing and Perception are critical for flexible manufacturing. Without robust knowledge of what is in the work area, where it is, and how it is behaving, it is not possible to operate safely without following rigidly-controlled trajectories. This prohibits human-robot collaboration, prevents dealing with unstructured environments, and adds cost through the need for fixturing of parts and addition of safety infrastructure. But in order to remain cost-competitive, perception must provide the same or better level of safety, localization, and identification as in current hard-automation systems. This motivates the need for performance metrics, standards, and calibration methods to quantify the capabilities of different sensors and to enable the right sensors to be selected for different applications.

  • Manipulation for Manufacturing Applications: The ability of humans to manipulate a wide range of objects with great dexterity and precision is what enables them to operate productively in the world. People can build things, take them apart, and determine many of their properties simply by touch. This is currently beyond the capabilities of even the most sophisticated robots. It is perhaps the biggest obstacle to moving robots into small and medium size enterprises. While there is growing research in force-based manipulation, there are no standards, performance measures, or metrics for manipulation, and no way for a potential user to decide if a device is dexterous enough to work in a particular application. The situation is more mature when larger-scale manipulation is the objective, such as moving material, packaging and palletizing, and loading and unloading trucks. Even there, however, there are few guidelines and most applications make use of purpose-built manipulators. In the micro- and nano-scale domain, manipulation is in its infancy and there is a great need for sensors that can measure in real time as well as manipulators and control algorithms.

  • Mobility for Manufacturing Applications: Mobile equipment is heavily used in manufacturing. There is a growing acceptance of either partially or fully autonomous mobile equipment in manufacturing. A major problem, however, is that (especially small) manufacturing facilities frequently operate with people and mobile equipment moving through the same cluttered and constantly-changing environment. Safety is of paramount concern, and standards are essential to reduce the potential for injury. Interoperability is also a concern, especially the ability to control multiple autonomous vehicles from different manufacturers and to mount different sensors from different vendors.

  • Autonomy for Manufacturing Applications: Autonomy encompasses all the computation and analysis of the state of the world that enables a system to behave intelligently. It includes organizing knowledge, planning actions based on goals and what is known about the world, and communicating with other agents (including people) to accomplish the required tasks. Planning is what enables a robot to take a priori and perceived information about the world and determine how to achieve its goals. In order to be able to plan effectively, the information about the world ("situation awareness") must be modeled in a useful way and constantly be kept up to date. There are a number of commonly-used representations for knowledge about the world, such as occupancy grids, topological models, and probabilistic representations, but there are no standards and no easy way of enabling one robot to inform another about what it knows. Since planning is heavily dependent on these representations, it becomes hard to develop planning algorithms that can work independently of a particular implementation. Also, an understanding of the nature of different tasks (task ontologies) is needed to enable automated planning, and the planning methods themselves need to be developed in a way that is not specific to a single application.

The initial projects, organized under these thrusts, are as follows:

Sensing and Perception for Manufacturing Applications

1. Safety of Human-Robot Collaboration Systems

Safe human-robot collaboration is widely seen as key to the future of robotics. When humans and robots can work together in the same space, a whole class of tasks becomes amenable to automation, ranging from collaborative assembly to parts and material handling and delivery.  Keeping humans safe requires the ability to monitor the work area and ensure that automation equipment is aware of potential danger soon enough to avoid it. Work on this project will address the following objectives:

  1. Develop the safety standards and performance measures to enable humans and robots to work together in the same space.

  2. Develop performance measures for sensors used to monitor the work area and ensure safety of people, robots, and vehicles. 
2. Perception in Unstructured Shop-Floor Environments

The goals of perception in manufacturing are to reduce the requirement of maintaining the positions of parts through fixturing, to enable automated systems to adapt to variations in parts, to enable in-process inspection, and to let robots work in small, less automated facilities where the environments are not rigidly structured. The shop floor perception project will address the following:

  1. Performance evaluation standards for perception algorithms for pose and force.
  2. Calibration and registration of sensors.

Manipulation for Manufacturing Application

1. Dexterous Manipulation for Autonomous Systems

Autonomous platforms currently have far less dexterity than people do, and industry practice is to custom-build manipulators for each task. If robots are to become more flexible and adaptable they will need highly capable manipulators with multiple degrees of freedom, and robust control to make use of these manipulators practical. The objectives for this project will include

  1. Develop performance measures for dexterous manipulation.

  2. Develop dynamic force measurements and force-based control of manipulation.

  3. Develop collaborative manipulation strategies for safe human-robot or robot-robot operations.

  4. Develop standards activities in the area of robotic performance.
2. Micro- and Nano-Manipulation for Manufacturing Applications

Manipulation at the micro and nano levels requires different strategies than at the macro scale and also the development of related sensing and actuation strategies to be able to control the manipulators. Standards are largely absent from this domain and measurements are typically made using cumbersome equipment not suited for manufacturing environments. There is a need for fast, easy to use sensors, reliable manipulators, and practical production scale-up if micro- and nano-scale manufacturing are to become widespread. This project will address the following objectives:

  1. Develop new measurement methods and sensors for micro- and nano-scale objects (distance, force, motion, physical properties)

  2. Enable scaling up from the micro- to the macro-scale

  3. Participate in standards activities related to micro-manufacturing

Mobility for Manufacturing Applications

1. Mobile Autonomous Vehicles

Autonomous mobile vehicles are becoming more common in manufacturing environments, and forklifts with autonomous capabilities are also making an appearance. One of the biggest barriers to this proliferation is safety. Mobile vehicles frequently must operate close to people and equipment, and this leads to many industrial accidents every year. This project will address the use of autonomous mobile vehicles through the following activities:

  1. Develop standards for safety of manufacturing vehicles.

  2. Develop methods to enable multiple sensors on a vehicle to inter-operate and provide combined information to operators or to a control system

  3. Develop standards to enable vehicles from different manufacturers to work cooperatively in the same work environment

Autonomy for Manufacturing Applications

1. Intelligent Planning and Modeling

The ability to understand the world and create plans to operate in it enables intelligent reasoning behavior and advanced automation. Planning frequently involves evaluating 'what-if' scenarios, which requires simulating the future based on present knowledge. This project will put simulation and planning on a more rigorous foundation by addressing the following objectives:

  1. Develop planning algorithms for manufacturing applications

  2. Develop standard ways of representing information that facilitate planning and can be easily ported to new applications

  3. Develop standard methods to validate simulation models to ensure that they accurately reflect the real world.

  4. Develop standard ways of seamlessly integrating information across real and virtual environments.

  5. Develop standard ways to rapidly build new simulations.

  6. Develop performance measures for the accuracy and completeness of planning systems.

How will teamwork be ensured?

The projects are all carried out entirely within the Intelligent Systems Division, although staff members are drawn from all groups within the Division. Most staff will work on multiple projects, and projects have specific ties where one project depends on another for data, capabilities, or expertise. Most projects will share a common test bed and participants will work to derive common scenarios to work towards The projects will hold regular project meetings and the program will hold periodic meetings with project leaders to ensure consistency and collaboration between projects.

What is the impact if successful?

U.S. industries will attain greater responsiveness, productivity, and higher quality through more widespread adoption of robotics and automation, which will be enabled through the innovations stimulated and guided by the performance metrics, test methods, and standards developed in this program. This will allow the U.S. to effectively compete in the global market. Once robots are more flexible, reliable, safer, and easier to install and interface, their applicability and use will increase tremendously, enabling many more manufacturers to reap their benefits. Robots that can immediately respond to new instructions, without losing time for calibrations and shop floor adjustments, will allow manufacturers to produce smaller lot sizes while taking advantage of the accuracy and quality of industrial robots. Robots that can be trusted to work alongside and assist humans will reduce injuries due to repetitive motions and strains as they offload the heavy and tedious work from humans without needing special barriers and restricted areas. Robots that can navigate around a dynamic shop floor without needing meticulous and specific programming can be incorporated into even the smallest manufacturing shop. Manufacturing at the nano/micro level (i.e, "factory on a chip"), construction, medical/bio, and other industries will achieve breakthroughs if the proper robot and automation capabilities are enabled.

In addition to enabling a flourishing of industries that take advantage of the next generation of robots, the program will stimulate the robotics and automation systems and peripherals industries in the United States. The bounty of creativity and innovation in leading-edge technologies within this country can re-invigorate the domestic robotics industry and create new affiliated industries.

Standards Strategy

Standards Landscape

The major player in standards for manufacturing robotics and automation is the Robotic Industries Association. They develop the U.S. robot standards and represent the U.S. in the ISO standards committees. Industrial robotic vehicle standards are developed by the Industrial Truck Standards Development Foundation, while ASTM is active in standards for sensors and measurements. ASTM also has committees that are developing standards for aerial, aquatic, and urban search and rescue applications (US&R). NIST has held a leadership position in E54.08 (Homeland Security Applications, Operational Equipment) and has been very successful in convening stakeholders and promulgating several performance standards for US&R. This experience will be leveraged for the manufacturing robotics strategy, regardless of the host standards body. The IEEE has started to show interest in robotics standards, especially for knowledge representation and autonomy[21], but is only just starting up relevant committees (led by NIST personnel). Most of the standards for robots are related to safety. Existing performance standards for robots are outdated and with the new requirements for safe robots and the desire for more sophisticated robot control and better precision, there is growing industry pull to update them. There is also a start to standards at the nano-scale, for example through IEC TC113 (nanotechnology standardization for electrical and electronic products and systems). Thus, while the CCC roadmap and other publications call out the need for standards and performance metrics, the standards community is in the earliest stages of responding, and in areas outside of safety, the program will have to make significant efforts to educate industry and establish standards working groups. Early results in this area will likely be in the form of best-practice documents rather than standards.

Top Standards Development Needs

The main standards development needs for the program include:

  • the adoption in the U.S. of the ISO 10218 industrial robot safety standards, parts 1 and 2, to enable greater flexibility and reduce costs when designing robot work cells, and to increase worker safety (FY14).

  • the completion of ISO Technical Specification 15066 on human-robot collaboration. This is needed by the end of 2013 when it will likely form the basis of a new standardization effort. This technical specification provides the requirements that will enable humans and robots to share the workspace. When made into a standard, it will truly enable the next generation of automation. The effort is led by NIST and members of the German ISO committee.

  • the development of a new U.S. robot safety standard, RIA/ANSI 15.06 harmonized with ISO 10218. The standard will include U.S.-specific requirements and be integrated with other safety standards and regulations (OSHA recognizes the ANSI standards) (FY14).

  • the addition of new capabilities to the ANSI/ITSDF B56.x standards to allow for using sensors on autonomous and semi-autonomous vehicles used in manufacturing. This will reduce accidents and improve performance of industrial vehicles (FY15).

  • the completion of performance standards (ASTM E57) for static pose measurement systems for manufacturing applications will reduce the need for parts fixturing and expand the range of applications for which robots can be used cost effectively (FY13).

  • the completion of performance standards (ASTM E57) for dynamic pose measurement systems for manufacturing applications to further speed up assembly and allow on-the-fly operations (FY15).

In addition, new efforts will be initiated on standards for micro- and nano-manufacturing. Meetings with SDOs will be held in 2011 to determine which one is most appropriate to host new standards activities. Standards will also be initiated for performance measures for robots and for dexterous manipulation (through RIA), and for aspects of autonomy such as knowledge representation for intelligent planning and decision making (through IEEE).

Anticipated Impact

There will be impact both near-term and far-term. The emerging next-generation safety standards are essential prerequisites for making the goal of more intelligent and collaborative robots a reality.   In the farther term because technologies are still in the early stages, consensus standards that measure how well a system meets specific performance targets and that help characterize a robotic component, subsystem or system can help end users match appropriate solutions with their application environments and tasks. Standard test methods also provide a concrete framework within which solution providers can improve their offerings and advance the technology capabilities for next generation robotics.

In the near-term, standards efforts already underway will have an impact particularly in the areas of safety and perception. The robot safety standards will be adopted and used worldwide, starting in 2011. In the U.S., the ISO 10218 standard must first be harmonized with the RIA/ANSI R15.06 standard, which is expected by the end of FY14. This must be done because OSHA follows the ANSI standard. There are some potential problems with acceptance of all the requirements of the ISO standard, particularly as they relate to intellectual property, which have to be resolved. The ability to let people work in close proximity to robots will increase productivity and enable small manufacturers to adopt the use of robots with lower costs (by about 2020). The autonomous manufacturing vehicle safety standards will help reduce the number of vehicle-related accidents and will foster the market for safety sensors. This will happen starting around 2013, but will continue as each of the standards are updated to allow new features. The need for standards for pose determination in manufacturing has been discussed with industry for a number of years, especially with regard to reducing the amount of fixturing needed in parts handling and assembly. Static pose standards will be adopted, initially by the automotive industry, starting around 2013, while dynamic pose standards can be expected in 2015. The introduction of the safety and pose measurement standards is expected to first impact primarily the automotive, aerospace, and food processing communities where the bulk of robots are used, but will expand into other industries and should help small manufacturers who can't afford the cost of specialized fixturing devices.

Current and Alternate Standards Strategy

The program participates actively in both existing standards committees and in helping to set up new standards working groups where a need is evident (determined mostly from discussions with industry and from documents like the CCC roadmap). The program will take on different roles for standards in different areas. For example, in the robot safety area, it will provide critical technical input, leadership of human-robot collaboration activities, and validation and conformance tests. In the measurement area, the program will provide leadership in convening the standards groups and will also develop conformance measures for the standards that are developed. In other, performance-related standards, the program will need to work with SDOs and manufacturers to initiate new standards activities.

The program will continue its active role in ISO and U.S. robot safety standards, including ISO 10218 parts one and two, the U.S. robot safety standard RIA/ANSI R15.06, and two related to industrial vehicles, ANSI/ITSDF B56.x and ISDO/DIS 13564-1. The program will continue leadership of measurement-related performance standards, including ASTM E57 standards development for 3D measurement systems and for non-contact pose measurement systems. The program will also develop conformance and validation tools for the standards to help speed adoption by industry.

Fit to Criteria for Selecting Standards Development Involvement

The standards address high priority needs for technology intensive manufacturing, and are therefore directly in line with the EL mission.  The standards reduce barriers to the adoption of robotics and automation and require NIST-specific measurement science to develop performance measures and metrics. As indicated above, the standards will benefit both large and small manufacturers in many sectors and will put U.S. manufacturers in a more competitive situation with respect to their foreign counterparts. The safety standards will be used by all manufacturers who make use of robots, and all will benefit from lower costs arising from the reduced infrastructure (floor space, fencing, interlocks, etc.) allowed by the new standards. Workers will also benefit from increased safety, especially from improvements to the autonomous manufacturing vehicle standards. Small and medium-sized manufacturers will benefit from both safety and performance standards by being able to adopt robots and automation systems at a lower cost and with greater flexibility and applicability to their needs.

There is significant and broad industry participation in all the standards efforts that are currently under way, even within the sensor performance standards that have only recently been started. There is a cross-section in all the committees of technology providers, end users, and systems integrators. Labor union representatives also participate in robot safety standard developments. There is also participation by other industries, but to a lesser extent. With the strong growth of robotics and automation in packaging and food processing, it can be expected that these groups will start to participate more in standards development.

The work is aligned well with planned program activities, and yet there is a need to extend the capabilities of staff in some areas to support aspects of standards that are new (e.g, in force-based control and dexterous manipulation). There will also be a need to augment the existing robot test bed to include a new generation "safe" robot with a highly dexterous manipulator. The resources and staff are sufficient to be successful, although demands from industry are greater than the program can address.

Actual Impact

  • ASTM E2807 - 11 Standard Specification for 3D Imaging Data Exchange, Version 1.0, describes a data file exchange format for three-dimensional (3D) imaging data. This standard was created with NIST leadership. Although only published in 2011, the standard has been adopted by the main 3D imaging system vendors, including FARO, InteliSum, Inovx, Kubit, Leica Geosystems, Optech, Pointools, Quantapoint, Riegl, Trimble, and Zoller+Fröhlich. The standard will be used both in the U.S. and internationally.

  • ISO 10218:2011 Robots and robotic devices -- Safety requirements for industrial robots.  NIST provided substantial technical contributions to this new standard. The standard will be adopted world-wide by robot manufacturers, integrators, and users and will reduce the costs of safety systems while enable greater flexibility in setting up robotic work cells, especially for human-robot interaction.

  • ANSI/ITSDF B56.5-2005 defines the safety requirements relating to the design, operation, and maintenance of unmanned automatic guided industrial vehicles and automated functions of manned industrial vehicles. NIST provided performance measures to enable the use of non-contact bumpers on the vehicles. This allows increased speeds of vehicle and reduces the risks of injury. The standard is widely used in the U.S by automotive, pharmaceutical, general manufacturing, chemical, and food and beverage industries.[22]

How will knowledge transfer be achieved?

Knowledge will be transferred through participation in standards activities, through conferences including the long-running series Performance Metrics for Intelligent Systems (PerMIS), through interaction with industry and academia, and through conference and journal publications. Some of the standards that result from the work are universally accepted within industry so improvements based on work in the program will have immediate impact. The program will leverage the Rapid Innovation and Competitiveness (RIC) Initiative to facilitate reaching out to research institutions in order to enhance U. S. technological innovation and connect academic capabilities with industrial needs. NIST is also uniquely positioned to effectively diffuse these research results on a broad basis in conjunction with the emerging technology deployment focus of the MEP and the proposed Advanced Manufacturing Technology Consortia (AMTech) Program. The work will also be transferred through PCAST-recommended public-private consortia and by providing access to NIST test bed facilities.

Outcomes:

This is a new program, but some related results have been obtained from predecessor programs. They include:

  • Establishing several CRADAS to transfer technology in the micro- and nano-manipulation area.

  • Patent disclosures were submitted and one of the patents was licensed by industry.

  • In the mobility area, results of NIST work have led to changes in the B56.5 safety standard for automatic guided vehicles that allow non-contact sensors. This has had significant impact in industry, resulting in the growth of a market for such sensors, and their use is extending to manned forklifts.

  • Other safety outcomes included a first draft of a technical specification (TS 15066 Robots and robotic devices — Collaborative robots) for the ISO 10218 standards (with the German ISO representatives).

  • NIST personnel also participated in drafting the ISO 10218 standards (parts 1 and 2), which were balloted in 2011 and have now been approved
    (ISO 10218-1:2011 Robots and robotic devices -- Safety requirements for industrial robots -- Part 1: Robots
    ISO 10218-2:2011 Robots and robotic devices -- Safety requirements for industrial robots -- Part 2: Robot systems and integration.)

  • Other standards outcomes include adoption by industry of the ASTM E57 data transfer standard and development of a first draft of a standard for static pose measurement systems (also under E57).

  • Program members have also been active in organizing and running competitions to test performance metrics. The virtual manufacturing automation conference is held annually, and a new perception challenge competition was inaugurated in 2011, jointly with Willow Garage, a robotics and computer vision startup.

Over the life of the program, the main expected outcomes include:

Safety

  • A Technical Specification on human-robot collaboration for the ISO 10218 Robot Safety Standard

  • Adoption of the ISO 10218 standards in the U.S.

  • Performance evaluation procedures for human detection that will enable verification of sensing systems and comparison between products

  • Measures to ensure force limits are not exceeded when humans work in contact with robots

Perception

  • A draft standard for static position and orientation measurement systems will address industry needs for locating parts without fixturing

  • Industry needs for calibration and sensor technology maturity measures will be developed through a workshop process. These measures are similar to the technology readiness levels used in the military and at NASA. They indicate the degree of maturity of a technology and its readiness for use in production systems.

Dexterous Manipulation

  • With industry input, a document (NIST-TN) will be prepared describing the need for dexterous, force-based manipulation in unstructured environments

Micro and Nano Manipulation

  • Measurement methods will be developed for micro- and nano-scale devices, including a system that can image the surfaces of micro/nano scale objects manipulated free of any substrate. Measurements and standards will include performance of micro- and nano-manipulation systems, design for manipulability, interfaces, coordinate systems and motions, safety, and vocabulary.

  • The mathematical framework will be developed for the gripping and assembly of complex micro scale devices to allow 3D devices to be built

Mobility (and Safety)

  • Measurements to support changes in the B56.x industrial vehicle safety standards will be carried out, particularly related to high-lift vehicles and visibility in manned vehicles

Planning and Modeling

  • Standard representations will be developed through the IEEE for world knowledge and plan knowledge to enable rapid re-tasking.

Recognition of EL:  

  • 2010 Bronze medal award for developing open source software

  • 2009 Government Open Source Conference Agency Award for Engaging Citizens

  • 2010 Best paper award from Standards Engineering Society at the World Standards Day Paper Competition


[1] PCAST, "Report to the President on Ensuring American Leadership in Advanced Manufacturing", June, 2011

[2] Ibid.

[3] EUROP: European Robotics Technology Platform, http://www.robotics-platform.eu/cms/index.php?idcat=41&idart=391, April, 2011.

[4]" A Roadmap for US Robotics: From Internet to Robotics," CCC and CRA (NSF-funded), 2009, http://www.us-robotics.us/.

[5] OMB-OSTP Science and Technology Priorities memo, July 21, 2010 http://www.whitehouse.gov/sites/default/files/microsites/ostp/fy12-budget-guidance-memo.pdf

[6] Executive Office of the President, December 2009, "A Framework for Revitalizing American Manufacturing" http://www.whitehouse.gov/sites/default/files/microsites/20091216-maunfacturing-framework.pdf

[7] http://www.roboticsbusinessreview.com/articles/newsletter_view/roadmapping-robotics-opportunities

[8] Gates, Bill. "A Robot in Every Home." Scientific American Jan. 2007: 58-65.

[9] "Robotics in Manufacturing Technology Roadmap," Produced by Energetics, Inc., Sponsored by U.S. DOE, October 2006.

[10] "Robotics in Manufacturing Technology Roadmap," Produced by Energetics, Inc., Sponsored by U.S. Department of Energy, October 2006.

[11] World Robotics 2006 statistics summarized by H. Christensen in presentation "'New' Applications in Industrial Robotics," RoboBusiness 2007, Boston, MA, May 2007.

[12] ibid

[13] Bekey, G. et al., "Panel Report on International Assessment of Research and Development in Robotics," Produced by the World Technology Evaluation Center, Sponsored by the NSF, NASA, and NIMIB, Jan. 2006.

[14] "Robotics in Manufacturing Technology Roadmap," Produced by Energetics, Inc., Sponsored by U.S. DOE, October 2006.

[15] Spooner, J., "Tech Standards Could Be Robotics' Road Map to Success," EWeek, June 21, 2006.

[16] "Robotics in Manufacturing Technology Roadmap," Produced by Energetics, Inc., Sponsored by U.S. Department of Energy, October 2006.

[17] In the NSF/CCC/CRA study, "Participants noted that research in robotics is rarely thoroughly evaluated and tested in well-defined, repeatable experiments"

[18] ISO 10218, ISO 13482, RIA/ANSI R15.06, ANSI/ITSDF B56.5.

[19] Messina, E., "Performance Standards for Urban Search & Rescue Robots: Enabling Deployment of New Tools for Responders," Defense Standardization Program Office Journal, November 2007.

[20] Dynamic Measurement and Control for Autonomous Manufacturing Workshop Draft Report, Sponsored by NIST, October 2007, Columbia MD.

[21] R. Madhavan, "RAS Standing Committee for Standards Activities—An Update on Recent Activities." IEEE Robotics and Automation Magazine, March 2011.

[22] http://en.wikipedia.org/wiki/Automated_guided_vehicle.

NDC0021_small
NIST robot testbed.

Start Date:

October 1, 2011

Lead Organizational Unit:

el
Contact

General Information:

Elena Messina
301 975 3510 Telephone

Mike Shneier
301 975 3421 Telephone

100 Bureau Drive, M/S 8230
Gaithersburg, MD 20899-8230