MOV.AI disrupts Autonomous Mobile Robot development with a Robotics Engine Platform that contains everything needed to quickly build, deploy and operate intelligent robots.
Digital twins are all the rage in intralogistics—but why?
With the rise of e-commerce and the need for fast fulfillment, warehouses are facing challenges in managing a growing number of SKUs. To meet these challenges, they employ automation at an ever-increasing rate, and warehouse environments are becoming increasingly complex.
Digital twins allow companies to visualize their warehouse operations virtually, without the need for shutdowns or trial-and-error approaches. By simulating different scenarios, companies can optimize floor plans, workflows, and other variables to enhance overall performance, improving efficiency by 20 to 25 percent. Digital twins offer a low-risk way to identify feasible solutions, reduce operating expenses, optimize equipment deployment, and gauge the potential impact of mechanization and automation.
All this is good and well, but when you dig deeper, you may notice an interesting phenomenon: one of the most critical elements of advanced warehouses—Autonomous Mobile Robots—often have a very superficial digital representation. What’s more, there is almost no talk of how an accurate digital representation of an AMR can improve robot development, speed up development and improve the fit to market needs.
In this article we discuss the importance of digital twins for AMRs and why AMR manufacturers should make sure they have a digital twin on hand.
While both digital twins and simulations involve creating virtual representations, the key difference lies in their purpose and connection to the physical world.
Digital twins are highly accurate representations of a physical entity.
Simulations, on the other hand, are computer-based models used to test and predict system behavior under different conditions. They are focused on exploring hypothetical scenarios, testing different strategies, and predicting outcomes based on predefined models and assumptions.
Simulations of complex environments can use digital twins to accurately test different hypotheses, software, or configurations before investing in building out the physical entity, be it a single AMR or a full blown warehouse.
Let’s dive deeper into AMR digital twins and their use throughout the AMR lifecycle
Advanced AMRs are complex systems that combine proprietary hardware, proprietary software and specialized components, while adhering to a growing number of standards. Roughly speaking, we can distinguish between AMR hardware, components (with their software), functional AMR software, operational or non-functional AMR software, and APIs.
Robots contain numerous interfaces and integrations between software, hardware components, and subsystems. To this are added the numerous integrations and interdependencies with the automation environment.
To validate and test AMRs, manufacturers need to simulate the software that brings it all together—the algorithms and software that dictate robot behavior given specific tasks and environment.
As all these elements are interdependent, it is more efficient to first run tests in a simulated environment using a digital twin of the robot.
Simulation plays a pivotal role in the development of autonomous mobile robots. It enables developers to test and evaluate the robot’s performance across diverse environments and scenarios before its physical deployment. By simulating the robot’s interactions, sensor data, and actuation, developers can identify weaknesses and fine-tune its functionalities, thereby ensuring optimal performance in real-world settings.
Robot validation occurs on different levels, and as such requires simulation of various aspects of the robot’s behavior and its environment. This is done using several types of digital twins:
A digital twin of the robot itself—hardware and components—and its physical environment.
Physics: Simulating the behavior of objects and the robot itself based on the laws of physics facilitates the validation of vehicle dynamics, collisions, slippage, friction, and reflections.
Actuation: Modeling the dynamics and kinematics of the robot’s actuators enables the simulation of precise movements based on control inputs, ensuring reliable and efficient actuation.
Sensors: Simulating the data generated by sensors in different simulated environments allows manufacturers to verify the accuracy and effectiveness of sensor inputs, and test the algorithms relying on sensor input.
Environment: Simulating diverse environments enables manufacturers to evaluate how the robot interacts with factors such as moving people, varying lighting conditions, and obstacles in its pathway.
A digital twin of robot management software:
Fleet management: Simulating different fleet sizes and testing higher-level fleet management aspects such as traffic management, task allocation, and analytics, ensures the seamless operation of a fleet of robots.
Task simulation: Simulating fleet-level tasks, such as inputs from a warehouse management system, allows manufacturers to validate the fleet’s ability to handle various scenarios efficiently.
A digital twin of the integration of AMRs with the greater automation environment
Traffic scenarios: Simulating different traffic scenarios helps manufacturers ensure that individual and/or fleet management software can avoid or recover from traffic conflicts, accounting for obstacles, human movement, and other fleet robots.
Digital twins are used by robot manufacturers to test the software algorithms and the holistics of AMR behavior under multiple test scenarios. Testing robot software using digital twins enables more granular testing earlier on in the process. Similar to the “Shift-Left” transformation in software development which was popularized several years ago, earlier testing of smaller processes reduces software development cost and time.
Below are some examples of the use of simulation and testing in robot development:
Warehouse navigation and obstacle avoidance: Simulation is used to test the AMR’s ability to navigate freely in a warehouse environment while avoiding obstacles. This ensures efficient and safe movement within complex operational spaces.
Pallet mover functionalities and success rates: Simulation enables the evaluation of pallet mover functionalities and assesses the success rates of pallet picking. This helps optimize the robotic system’s performance and overall productivity in handling specific tasks.
Recovery feature development: Simulation is employed to test the functionality and effectiveness of the recovery feature. By simulating various recovery scenarios, developers can fine-tune the feature to enhance performance and reliability.
Speed up development and reduce costs: Simulation allows manufacturers to conduct extensive testing without relying solely on costly physical hardware.
Efficient design decisions: Through simulation, manufacturers can make informed design decisions, such as optimizing sensor positions and orientations. This ensures that the robot captures sufficient data to effectively support its intended use cases.
Evaluate performance in different environments: Simulation empowers manufacturers to assess their robot’s performance across various scenarios, ensuring robustness and adaptability to different environments.
Sales enablement: Manufacturers can leverage simulation to showcase their AMRs in a simulated environment, effectively shortening sales cycles to automation integrators and end customers.
Simulation continues to play a vital role during the deployment and ongoing operation of AMRs, allowing automation integrators to assess fleet performance in a simulated world before real-world deployment. By utilizing simulation, integrators can identify and resolve potential issues, optimizing fleet setup and achieving desired performance indicators.
During fleet deployment, simulation focuses more on the overall behavior and performance of the fleet rather than individual robots. Running simulations with varying fleet sizes and faster-than-real-time speed helps measure KPIs, adjust setups accordingly, and ensure successful integration with broader automation solutions.
The digital twins used during robot development are also used during deployment and operation, only for different types of simulation and test scenarios. During manufacturing, tests are designed to test the robot and software functionality in general. During deployment and operation, tests are designed to validate site and task scenarios.
During deployment and operation, integrators use digital twins of:
It is important to note that as fleet size increases, some computationally intensive processes like 3D photorealistic rendering, localization, and navigation may need to be simplified to ensure the simulation can handle larger fleet sizes without compromising performance.
Site setup optimization: Simulation enables integrators to optimize site setup and fleet requirements in order to meet performance KPIs. The simulations consider factors such as tasks to be performed and the allocation of logic, traffic, physical site design, as well as achieving operational stability and throughput aligned with the broader automation solution requirements.
Support for site update setup decisions: Integrators can test the impact of changes to the site setup on fleet KPIs, ensuring effective decision-making and minimizing potential risks.
Reduced deployment costs: Simulation allows integrators to remotely set up a significant portion of the deployment, minimizing on-site time and reducing the effort required for fine-tuning.
Performing ongoing changes and optimization: during ongoing operation, required changes can be seen as a small deployment. For example, a change in a business process such as a plan for a new pick drop can be validated in a digital twin. Tests can be run to ensure the change provides the required KPIs and does not compromise other processes.
Support sales efforts: Simulations using a digital twin of a customer’s real environment and recorded data can support sales efforts. By showcasing a simulated fleet operation that meets the solution’s KPIs, automation integrators can effectively demonstrate the benefits of their solution to end customers using visually impressive, photorealistic simulations.
Digital twins play a pivotal role in the development and deployment of autonomous mobile robots. They empower manufacturers and automation integrators to optimize performance, reduce costs, and enhance efficiency throughout the entire lifecycle of AMRs. By leveraging simulation engines and embracing realistic scenarios, stakeholders can unlock the full potential of these robots, leading to improved productivity and operational success in various industries.
The utilization of autonomous mobile robots (AMRs) has witnessed widespread adoption across various industries such as manufacturing, logistics, and healthcare. As these industries embrace the potential of AMRs, it becomes crucial to navigate the complexities associated with their development, deployment, and operation. Simulation and digital twins have emerged as a powerful tool for ensuring the seamless functionality and efficiency of these robots in real-world scenarios. By leveraging digital twins, developers can proactively identify and address potential issues, significantly reducing costs and development time.
*By providing my details I agree to the MOV.AI Terms and conditions and Privacy policy. I also agree to receive emails from MOV.AIand I understand that I may opt out of MOV.AI subscriptions at any time.
