George Chowdhury, ABI Research Industry Analyst
Robots are becoming a staple across various sectors, automating repetitive and dangerous tasks within operations. ABI Research forecasts that annual robotic shipments—spanning industrial, collaborative, mobile, and exoskeletons—will increase from 1.38 million units in 2024 to nearly 5 million units by 2030. Designing and deploying robots can be convoluted, requiring frequent testing of various hardware and software. Simulating robotics deployments simplifies these processes, empowering engineers to test their solutions in a virtual environment. Robotics simulation tools enable companies to safely verify the robot’s component compatibility and capabilities before physically deploying it in a smart factory or warehouse.
A recent ABB survey found that 43% of U.S. companies plan to adopt robotics and automation to make their supply chains more resilient. The magic behind each of these robots functioning optimally and safely is rigorous testing before deployment. To make this happen, robotics stakeholders can leverage advanced simulation technologies like digital twins and Virtual Commissioning (VC). Highlighting the growing demand for these tools, ABI Research forecasts that robotics simulation software revenue will grow from US$584 million in 2023 to US$1.39 billion by 2030.
What Is Robotics Simulation?
Robotics simulation involves emulating a robot's hardware and software functionality in a virtual environment. This digital realm is populated by virtual representations of the robot and the environmental elements it encounters daily. By using robotics simulation software, engineers can test a robot’s design and responsiveness without needing to deploy it physically on the plant or warehouse floor. Furthermore, advanced robotics simulation software generates large volumes of synthetic data, which can be harnessed to reduce robot training time and enhance Artificial Intelligence (AI) models (learn more in the free whitepaper, Innovation In The Era Of Artificial Intelligence: Key Takeaways From NVIDIA's GTC 2024).
Related Insights
The Watershed Moment for Humanoid Robotics? (IN-7148)
A Revolutionary Charging Method for Mobile Robots (IN-7294)
Robotics Simulation Improves Deployment Confidence
Robotics simulations were once confined to academia and Research and Development (R&D), but the landscape has progressed. Advanced simulator technologies like digital twins, virtual commissioning, and hardware-in-the-loop are now considered essential for successful robotic deployments. These solutions utilize various Internet of Things (IoT) and sensor data within a facility to provide real-time insights into a hypothetical robot deployment. Synthetic data and generative AI further enhance simulation outcomes by supporting robot training and learning algorithms.
Simulating a robot’s capabilities in a virtual environment is safer and more efficient than testing a physical robot. For example, physically testing a robotic arm at a car manufacturing assembly line can disrupt the production process and risk operational downtime. On the other hand, a robotics simulator allows this testing to be done digitally and without the physical robot. Similarly, verifying a virtual robot’s performance assures engineers that the real-life counterpart will work properly and safely once deployed.
The three pillars of robotics simulation include digital twins, virtual commissioning, and hardware-in-the-loop.
- Robotic Digital Twin: A digital twin is a virtual replica of a robot that mimics the functionality and real-time data generation of the physical robot. From there, engineers can simulate how the robot will perform in the field. Crucially, the digital twin ensures the robot's design is optimized for safety, speed, strength, or other desired attributes. Notable vendors in the robotics digital twin space include NVIDIA, Siemens, Dassault Systèmes, ABB, etc.
- Virtual Commissioning: System Integrators (SIs) increasingly use Virtual Commissioning (VC) for robots in the warehousing and manufacturing sectors. VC gauges synergies between Programmable Logic Controllers (PLCs), Human-Machine Interfaces (HMIs), and robot controllers. This way, VC ensures that the robot hardware functions as engineers intend. Ultimately, VC minimizes downtime for new robot deployments, proves Return on Investment (ROI) to stakeholders, reduces the risk of physical damage upon deployment, and accelerates robotics innovation through demonstrable value.
- Hardware-in-the-Loop: Typically the final stage of VC, Hardware-in-the-Loop (HIL) involves physical hardware being put through a simulator to test controllers. In this process, engineers validate a robot’s performance by sending commands from dummy PLCs to a robotic arm controller or other robotic form factors. Alternatively, engineers may send commands from dummy Warehouse Execution Systems (WESs)/Manufacturing Execution Systems (MESs) to PLCs. HIL simulation’s biggest benefit lies in the time savings and simplicity of testing a robot’s controller without requiring the entire system.
Advanced simulation tools like those listed above streamline the design and deployment processes of robots. Ultimately, virtual testing of a robotic solution reduces Time to Market (TTM) for both its manufacturer and the enterprise using it.
Migrating Robotics Simulations to the Cloud
My team and I assert that cloud computing has immense potential to impact robotics simulation. As computational and Capital Expenditure (CAPEX) requirements for robotics continue to increase, it’s vital to offload computational power. Already, vendors like ABB and NVIDIA are setting a precedent for cloud-based robotics simulation with RobotStudio and Isaac Sim, respectively.
Cloud robotics simulation kills two birds with one stone. First, it enables companies to outsource the computational cost of training Machine Learning (ML) models for robotics applications. This reduces the strain on the local server that already handles vast sums of data every day. Second, hosting robotics simulators in the cloud unlocks the ability to disperse tested software to multiple sites simultaneously via cloud containers, streamlining deployments and reducing costs.
Computational speed and connectivity latency are more limited for cloud-native simulations than on-premises simulations. For this reason, offline simulations or VC in the cloud are not recommended for mission-critical use cases. ABI Research observes an up-and-coming computational solution called “cloud bursting” as an effective way to strike a latency/computational balance. With cloud bursting, the robotics simulator will leverage cloud resources when on-premises hardware is overwhelmed. Amazon Web Services (AWS) and Microsoft Azure—both already active in the robotics simulation space—are two cloud hosts well-suited for cloud bursting.
The logistics and warehousing sectors are prime opportunities for cloud bursting in robotics simulations. Applications such as Autonomous Mobile Robots (AMRs), bin picking, and inspection involve expensive hardware, but they can afford greater latency than their counterparts in industry and manufacturing.
Robotics Simulation and Digital Transformation
Advanced robotics simulation offers a huge leap in product design and deployment capabilities over traditional brute force prototyping. The streamlining and cost-savings benefits are hard to ignore. When digital twins and VC are combined with Artificial Intelligence (AI) synthetic data and cloud computing, industrial robotics stakeholders have an unprecedented level of deployment flexibility and optimization at their disposal. These solutions bring manufacturing and warehousing facilities one step closer to fully automated operations. For these reasons, you can’t have the digital transformation discussion without mentioning robotics simulation.
This technical overview comes from ABI Research’s Robotics Software: Simulation, Virtual Commissioning, and Emerging Innovation report. For a deeper look at the emerging practices, key vendors, and innovative use cases of robotics simulation, download the report.
