Performance Simulation Project

Automated Guided Vehicles (AGVs) and forklifts represent integral components of today's manufacturing processes. They are widely used on factory floors for intra-factory transport of goods between conveyors and assembly sections, parts and frame movements, and truck-trailer loading/unloading. Automating these systems to operate in unstructured environments presents an exciting area of current research in robotics and automation. This project provides researchers and industries with the necessary tools, standards, performance metrics, and infrastructure to lower the entry barrier for involvement in this research area.  The project collaborates with the simulation community, AGV developers, researchers, and end-users to create a combined real/virtual system that will advance the state of autonomous robotic technology by allowing for the safe and efficient development of collaborative systems that work in unstructured environments.

According to Bishop Consulting’s  report on AGV Industry Next-Generation Technology Priorities , “In the eyes of the system vendors, the most prominent technology development area is in moving from today’s AGVs, which require highly structured environments and reference markers installed throughout the plant, to operating in less structured or unstructured environments. In fact, the preparation required to site and install these reference markers are a significant portion of the system cost… In implementing AGVs for unstructured environments, another need is for sophisticated simulation capability… no capability currently exists to simulate AGVs for unstructured environments”.

The need for operation in unstructured environments exists in the full spectrum of manufacturing companies. From small- and medium-sized manufacturers whose entire shop floor is too unstructured and dynamic for today’s AGV systems, to large-scale manufactures that have unstructured “buffer-zones” where temporary storage occurs. This project applies techniques developed in the Intelligent Systems Division for Urban Search and Rescue (USAR) and bomb disposal robots, which are producing consensus-based performance test method standards through ASTM International, to the AGV community in order to develop performance test method standards for AGVs. These techniques have been formalized in the System- and Component-evaluation in Operationally Relevant Environments (SCORE) framework. The SCORE framework provides for the evaluation of a system at the component level, the system level, and in operationally-relevant environments. Following a similar requirements-based process, we will convene a panel of industry representatives to catalog envisioned technological advancements that will necessitate new performance requirements.   Test methods will be developed to quantitatively and reproducibly measure how well systems perform against the requirements.

Unfortunately, the traditional entry barrier into this research area is quite high. Researchers need an extensive physical environment, robotic hardware, and knowledge in research areas ranging from mobility and mapping to behavior generation and scheduling. An accepted approach to lowering this entry barrier is through the use of simulation systems and open source software.

We therefore take the novel approach of developing and refining test methods in simulation in a virtual factory. We will then transition them to easily realizable physical artifacts that remain coupled to the simulation environment. The reason for the coupling is that it is anticipated that several of the test methods will involve issues with human-machine interaction. By utilizing a coupled real/virtual test environment, system performance around humans can be quantified without posing any actual risk to human participants. Knowledge gained during this project will also be applied to develop performance metrics for the safe operation of robotic arms around humans as part of the Mobility and Manipulation Project.

We will also take the novel approach of validating our performance metrics through competitions. IEEE and NIST are cosponsoring an international Virtual Manufacturing and Automation Competition (VMAC) that will use tentative performance standards as the judging criteria.