Autonomous mobile robots have arrived. Autonomous mobile robots (AMRs) are capable of running safely and efficiently without human supervision. But can AMRs work for you? This article talks about some of the examples of where and how to us AMRs in the biologics laboratory.
What is the difference between and automated guided vehicle (AGV) and an autonomous mobile robot (AMR)?
If you think that autonomous mobile robots (AMRs) are the same as automated guided vehicles (AGVs), here is some information to change that view.
Automated Guided Vehicle (AGV) Background
AGVs are systems that follow a preset guided route on their path from one point to another. The guide route can be anything from rails, RFID tags, or even a line of magnetic paint that the vehicle follows. When the route is blocked, AGVs can slow or stop. Advanced AGV systems can move along an alternative guided route if the first option is unavailable.However, AGVs are not robots as demonstrated by the need for preset routes.
Autonomous Mobile Robot (AMR) Background
Autonomous mobile robots (AMRs) act as a true robot following the Sense-Plan-Act paradigm. AMRs must have at least one sensor that can perceive the world around them. Most AMRs use multiple light detection and ranging (lidar) systems which provide precise and accurate line of sight information about physical obstacles that enables the AMR plan the route of travel between or around any obstruction and then act on that plan. Lidar is on at all times so that if a physical route of travel changes, the AMR can respond in real time by stopping or redirecting the system as appropriate. Unlike AGVs, AMRs do not require preset routes. As a result, they are ideal in a constantly changing environment.
Links to other sites that describe differences between AGVs and AMRs
- https://www.mobile-industrial-robots.com/en/insights/get-started-with-amrs/agv-vs-amr-whats-the-difference/
- https://fetchrobotics.com/fetch-robotics-blog/amrs-vs-agvs-whats-the-difference/
- https://waypointrobotics.com/blog/amr-vs-agv/
- https://ottomotors.com/resources/info/agv-vs-amr
- https://www.agvnetwork.com/agv-vs-amr
What are autonomous mobile robots (AMRs) good for?
AMRs can navigate from one location to another without requiring any human intervention. Their ability to detect obstacles allows them to modify their navigation route if something or someone gets in their way. AMRs are already used in warehouses. If you work with automation in laboratory setting, there are multiple reasons to consider using an AMRs including:
- Move samples between storage locations
- Transfer samples between automated processing equipment and analytical instruments
- Deliver or transfer samples between equipment and scientists
Laboratory robotics and automation today
Automated equipment is programmed in most labs to perform a specific processing function, workflow, or analytic process. Scientists are trained to walk up to a liquid handling robot or other automated system and add samples and consumables before starting the processing allowing the scientists to do other things in the lab while the automated equipment is functioning. After the system completes the task, the scientist is then required to remove the processed sample or any consumables that need to be discarded. They ideally also clean the system for the next scientist who needs it.
Laboratory robotics and automation currently suffer from the following 3 challenges:
- Dependency on Nike Automation
- Limited Operational Windows
- Co-localization of physically integrated systems
Scientists who work in highly automated labs often find themselves performing Nike Automation
Manual tasks that keep automation operational is often referred to as “Nike Automation”. Some examples of Nike Automation includes:
- loading/unloading a system with consumables and reagents,
- transferring samples and products between systems, or
- planned manual interventions in robust processes to add chilled reagents or incubate samples.
Even with the productivity hit of Nike Automation tasks, most laboratory robotics and automation result in such substantial net gains in productivity, precision, and accuracy. However, the existence of AMRs and the ability to operate them in a lab setting is starting to challenge that paradigm.
Most automated equipment only operates in windows of time when scientists are in the lab (8 hrs/day and 5 days/week)
Since most laboratory robotics and automation require some form of Nike Automation, the equipment is usually only running when scientists are present for the beginning and end of the run limiting those automated processes to <8 hrs/day and <5 days/week.
You can extend the window of time of operation by split shifts or allowing systems to run to completion after a scientist has gone home for the day. However, that is only possible if the samples generated do not need to be transferred to refrigerated or frozen storage immediately after processing. Even with this extension of time there are many hours in a day where the systems sit completely idle.
Physical integrations of robotics and automation require co-localization of equipment and very large labs
One method used to create increasingly complex laboratory automation processes is to co-locate processing equipment and adding robot arms or shuttles to enable fully automated transfers of material between the systems.
The 3 primary challenge with these integrations is:
- All equipment in an integrations becomes dedicated to the integration with limited functionality outside the integration.
- Physically integrated systems take up a lot of space.
- Physical integrations presuppose the order of automated actions requiring significant work to modify a pre-existing integration.
In the end larger integrated systems tend to have the same challenges that stand alone systems have, they are just capable of automating more of a workflow. These may make sense for more dedicated workflows where the sole purpose of device is for that workflow. In general though these integrations come with a reduction in flexibility for laboratory set up and the process/workflow.
AMR enabled Laboratory Robotics and Automation
Autonomous mobile robots equipped with robot arms and sample storage enable us to address all 3 of the major challenges for laboratory robotics and automation today:
- Replacement of Nike Automation with autonomous mobile robot processes
- Expansion of robotics operational windows to occur on evenings and weekends
- Limiting physical integrations to only those workflows that require it
Goodbye Nike Automation and hello to new process bottlenecks
The ability to schedule the delivery and retrieval of samples, reagents and consumables changes the game when it comes to the need for Nike Automation. Even if there is still the need for some manual work on the part of the scientist before or after an automated workflow occurs, the need for delays in the middle of workflows should be all but extinct within a few years of implementing AMR.
Of course the ability deliver and retrieve samples also means the need for places to retrieve those samples from and deliver those samples to. Autonomous mobile robots are limited for biologic processes if they are not paired with automated storage.
Goodbye robot work weeks… Hello just in time robotics
AMRs will allow robust automated processes to occur on a timeline that fits a scientist’s needs. Imagine a future where the samples and data generated by laboratory automation are at your fingertips exactly when you need them.
The advent of just in time robotics not only has a large impact on a scientist’s productivity; but also on their work-life balance. This is especially true for automated cell culture experiments that currently require scientists to work on weekends to keep bioreactors fed and sampled as appropriate.Replacing physical integrations with flexible orchestrated robotics
Replacing physical integrations with flexible orchestrated robotics
Eliminating the need to physically integrate all robotic processes you want in contiguous workflows has an outsized impact on automation in three very important ways:
- Decrease the number of physically integrated systems and therefore the average size of automated systems
- Increase the number and variety of automated systems that can be connected in workflows leveraging AMR transport
- Simplify the path to change existing workflows
The first simplifies options for automation in older labs.
The second enables addition of multiple automated workflows to the existing robotics orchestrations. Therefore resulting in rapid expansion of biologics laboratory workflows.
The third allows workflows and processes to change as the science evolves without requiring massive investment or down time.
Process design and optimization can now follow an agile methodology instead of the more traditional waterfall. Waterfall development result in highly tiered and linear projects. Agile development enables both flexible and parallel tracked projects. The waterfall to agile transition results in changes happening organically and often with a better final product.
What do I need to do to be ready for Autonomous Mobile Robotics?
Below is the list of Robotics and Informatics capabilities you will need. Obviously, it is likely that you already have many of the things on the below list:
- Automated Storage (Frozen, Refrigerated, Incubated)
- Standardized Automation Consumables (Automation tubes)
- Automated Consumable handling
- Liquid handling Robots
- Automation Friendly Analytics
- Sample Management Software
- Automation Scheduling Software
- Data Extraction, Transformation, and Loading (ETL) capabilities
- Automatic Doors (for AMR access of labs)
It is still early days, but we are more impressed every day with what we are accomplishing with them.
