In response to the surging use of autonomous mobile robots (AMRs), also called industrial mobile robots, in Industry 4.0 operations, the Association for Advancing Automation (A3), together with the American National Standards Institute (ANSI), recently released the second increment of its safety standard for AMRs: ANSI/A3 R15.08-2, which details the requirements for integrating, configuring, and customizing an AMR or fleet of AMRs into a site. An essential requirement is the performance of a risk assessment per ANSI/ISO 12100 or ANSI B11.0. The new standard complements the previously released R15.08-1 that focused on the safe design and integration of AMRs.
The R15.08 series of standards builds on the earlier ANSI/ Industrial Truck Standards Development Foundation (ITSDF) B56.5 safety standard for automated guided industrial vehicles (AGVs). The newer standard recognises three classes of AMRs based on the inclusion of specific functions and features.
This article briefly compares AMRs and AGVs and ANSI/ITSDF B56.5 and International Standards Organization (ISO) 3691-4 versus ANSI/A3 R15.08. It then reviews the risk assessment strategies outlined in ANSI/International Standards Organization (ISO) 12100 and ANSI B11.0, how they relate to AMRs, and how they are integrated into R15.08-2.
Next, it reviews the three classes of AMRs defined in R15.08-2 before closing with a presentation of practical considerations for AMR integration, including how to implement mapping and commissioning, how to manage fleets of AMRs, and how to navigate new opportunities for virtual commissioning using simulation and digital twins using examples from Omron Automation and Siemens.
AGVs can travel only along a predetermined and marked path. They have no independent navigation capabilities. They stop if they arrive at an obstacle and wait for it to be removed before proceeding along the fixed path. AMRs include independent navigation systems and can change paths and move around obstacles (Figure 1). Because of these differences, AGVs are better suited for relatively stable and unchanging environments, while AMRs support more flexible and scalable deployments like those needed in Industry 4.0 operations.
Standard evolution
Some AMR standards have evolved from previously developed standards for AGVs and stationary robots. For example, EN 1525:1997 was developed for AGVs and was subsequently applied to AMRs without modification. The newer ISO 3691-4 standard covers AGVs and has sections dedicated to AMRs.
ANSI/ITSDF B56.5 is a Safety Standard for Guided Industrial Vehicles, unmanned guided industrial vehicles, and the automated functions of manned industrial vehicles; it does not cover AMRs. The newer ANSI/RIA R15.08 is a safety standard for the use of AMRs in industrial environments. It’s based on and expanded from the R15.06 standard for safely using stationary robotic arms.
Another important standard is EN ISO 13849, which defines the safety performance levels (PLs) for various types of equipment. There are five levels, from PLa to PLe, with increasingly stringent requirements. AGV and AMR makers must reach PLd safety that ensures continuous safe operation in the event of a single fault, i.e., by using redundant systems.
ANSI/A3 R15.08-2 requires a risk assessment for integrating and deploying AMRs. The risk assessments defined by ISO 12100 and ANSI B11.0-2010 are very similar, though not identical. ISO 12100 targets original equipment manufacturers, whereas ANSI B11.0 focuses more on machinery and end-user safety. The basics of risk assessment are similar for both standards.
Risk assessment
A risk assessment is a highly structured analysis to arrive at an acceptable level of risk. It recognises that no system or environment is perfect; inherent risks can be managed but not eliminated. It begins by determining the limits of the machine’s operation and identifies hazards that can arise if the machine operates near or outside of those limits.
Next is risk estimation, which looks at the likely severity of harm from each hazard and the probability of its occurrence. A very severe hazard with a low likelihood of occurrence may receive a similar ranking as a hazard with a less severe outcome that’s more likely to occur. All identified risks are evaluated and ranked to prioritise risk reduction efforts. Risk assessment can be an iterative process, identifying the most severe risks and reducing their probability of occurrence and/or the severity of their outcome until an acceptable level of residual risk has been achieved (Figure 2).
AMR classes
R15.08 recognises three types of AMRs:
Type A: AMR platform only. In contrast with AGVs, type A AMRs can function as independent systems without requiring environmental changes. They can include optional features like a battery management system, the ability to independently locate a charger and recharge its battery, the ability to integrate with centralised fleet management software, etc. Type A AMRs are most often used to move materials around a factory or warehouse.
Type B: A type A AMR with the addition of a passive or active attachment that is not a manipulator (Figure 3). Typical attachments include conveyors, roller tables, fixed or removable totes, lifting devices, vision systems, weighing stations, etc. Type B AMRs can be used for more complex logistics tasks. Vision systems can be used for product inspections and identification, weighing (or estimating the number of) parts, and so on.
Type C: A type A AMR with the addition of a manipulator. The manipulator can be a robotic arm with three or more axes of movement. Type C AMRs can be designed to function as collaborative robots (cobots) working alongside humans. They can also be machine attendants, perform pick and place operations, complete complex inspection tasks, do harvesting and weeding in agricultural settings, etc. Some designs can move from place to place and perform different tasks at each station.
Commissioning, mapping, and following the lights
All three types of AMRs are designed to simplify deployment. Compared with AGVs that require extensive infrastructure installation, no construction is necessary for AMR deployment, and programming needs can be minimal. Basic commissioning is a four-step process (Figure 4):
- The AMR is delivered with all the needed software installed; the first task is to install and charge the battery
- Mapping is critical and can be manually or automatically implemented. For manual mapping, a technician controls the AMR and takes it around the facility so it can learn about the environment. Laser-guided AMRs can automatically scan up to 1000 square feet per minute to create maps capturing all the features in the immediate area and wirelessly send the resulting map to a central computer. In both cases, maps can be customised with virtual routes and forbidden lines for safe operations and can be shared across fleets of AMRs
- Setting goals includes the identification of pick-up and drop-off locations
- Task assignment is the final step and includes scheduling and coordination of the various AMRs in the fleet and integration with Enterprise Resource Planning (ERP), the Manufacturing Execution System (MES), and the Warehouse Management System (WMS)
In addition to mapping a facility using laser scanning, some Omron AMRs use a camera to detect and plot the location of overhead lights. It creates and overlays a “light map” with the standard “floor map.”
Laser localisation can tolerate changing environments on the floor up to a point. Suppose over 80% of the features change, for example, on a shipping dock where pallets or rolling carts constantly change location. In that case, laser localisation is less useful, and adding the light map increases the reliability of navigation. Using the light map also enables AMRs to more easily navigate across wide-open areas in large facilities.
Managing robot fleets
Effective management of robot fleets can multiply the benefits of using AMRs. It can support centralised control and coordinated operation of mixed types of AMRs and provide the data and analytics needed to maximise operational efficiencies. Some common features of AMR fleet management systems include:
Optimised task assignments are based on the capabilities of each robot in the fleet, their current locations, and anticipation of where their next assignment will be located.
Traffic management includes scheduling pick-up and drop-off locations and times for maximum efficiency and notifying robots of destination changes or new obstacles, enabling them to recalculate their path for maximum efficiency and safety.
Charge management tracks the battery charge level of each robot in the fleet, enabling proactive charging and maximum uptime.
Coordinated software updates across the fleet to ensure the latest version is available for each type of robot.
Enterprise integration connects the fleet management software to ERP, MES, and WMS systems so jobs can be allocated and scheduled automatically to the fleet in real-time.
Virtual commissioning
A combination of digital twins and simulation software enables virtual commissioning. In this case, a digital twin is a virtual representation of an AMR. Digital twins can be used to virtually validate the performance of individual AMRs and fleets of AMRs. Virtual commissioning uses robotics simulation software to combine the digital twins of AMRs with a digital twin of the surrounding environment (Figure 5).
AMR virtual commissioning can also be used to integrate and coordinate the operation of robots from several manufacturers. During the virtual commissioning process, engineers can quickly and efficiently create multiple scenarios to verify the proper functioning of the entire system, not just isolated AMRs.
Virtual safety testing and debugging can also be implemented with digital twins and simulation. Virtual AMRs can be subjected to anomalous situations to test various contingencies and ensure the proper functioning of safety protocols.
The ability to implement virtual debugging can speed up the deployment of AMR fleets. Debugging fleets of physical AMRs after deployment is challenging and time-consuming. It involves work stoppages and negatively impacts the productivity of the facility. There are no work stoppages with virtual debugging, and users are assured that the AMRs will perform as expected in the real world.
Conclusion
AMR deployments are becoming increasingly prevalent in a wide range of Industry 4.0 installations. The standards landscape for AMRs is evolving to address requirements for safely and efficiently integrating, configuring, and customizing an AMR or AMR fleet into a site. A risk assessment performance is a key requirement within the new standards in accordance with ANSI and ISO standards. The tools for AMR commissioning are also evolving with the emergence of virtual commissioning using digital twins and simulation.
This was the first of a two-part series and focused on the implications of the recently released R15.08-2 standard regarding safety, risk assessment, and commissioning of AMRs. The second article is written in anticipation of R15.08-3, which is currently under development and will address the topic of sensor fusion in AMRs.
Author: Jeff Shepard, DigiKey contributor
Source: DigiKey
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by commenting below or visiting our LinkedIn page.
