
Conquering Robot Controller Design Challenges with Hypervisor Technology
A couple of months ago, while on a business trip in the U.S., I used an app that provides an eSIM service. With just a few clicks, the app allowed me to add a data plan on top of my existing one. In the past, without this virtual SIM card, I had to desperately search for a paperclip to pop out the SIM tray and switch between data plans while traveling. Or even use a second phone. With the virtual SIM card, I can easily switch between different data plans and keep multiple plans on a single phone without them interfering with each other. The virtualization technology I discuss here offers a similar advantage: It enables the consolidation of multiple workloads running on different operating systems onto a single system. This approach not only eliminates the need for additional systems and complex cabling but also facilitates more cost- and space-efficient designs. However, the question remains: Why do robotics customers require virtualization technology today, and what specific benefits does it offer?
The Evolution of the Modern Manufacturing Industry
As the manufacturing industry evolves, the focus is increasingly shifting to collaborative robots, or "cobots." Unlike traditional industrial robots, which operate in isolation, cobots are designed to work alongside humans, enhancing both efficiency and flexibility on the production floor. However, while the benefits of cobots are clear, designing their controller systems for efficient and secure operation is challenging. The primary hurdle lies not in the availability of technology, but in the process of integrating various components onto a single platform—in a way that reduces power consumption, cost, and design complexity (IoT Virtualization Jump-Starts Collaborative Robots, Brandon Lewis, 2022).
While many core technologies—such as AI and machine vision for visual perception, multicore processors for high-performance computing, and soft PLC software for process monitoring and control—are already available, the challenge is integrating these diverse components into a cohesive system. The complexity of cobot design lies in the need to balance performance with power efficiency, minimize hardware costs, and simplify the overall system architecture. This integration must also prevent any interference between applications running on general-purpose operating systems (GPOS) and real-time operating systems (RTOS).
The Role of Hypervisors in Workload Consolidation
This is where hypervisors and virtualization technology come into play. Traditionally, robotic systems have relied on multiple dedicated controllers to handle different tasks, from sensor data processing to motor control. However, this approach increases the overall hardware footprint, power consumption, and complexity of systems. By using a real-time hypervisor, robotics customers can consolidate multiple workloads onto a single high-performance hardware platform.
Hypervisors manage different operating systems, shared memory, and system events to ensure all workloads on a system remain isolated while still receiving the resources they need. This maximizes resource utilization and enables parallel real-time processing of deterministic and AI workloads without interference. This reduces the number of physical systems required and streamlines system management, making it easier to scale and adapt to new production requirements.
Integrating AI Vision and Motion Control on a Unified Hardware Platform
Moreover, the evolution of manufacturing increasingly relies on AI-powered vision systems for robotics. Cobots equipped with high-resolution stereoscopic cameras for 3D vision and AI algorithms can perform complex tasks, such as quality inspection, object recognition, and precision assembly, all in real-time. When combined with artificial intelligence decision-making capabilities, these collaborative robots can quickly learn and optimize methods for task execution, achieving higher efficiency in task completion (Collaborative Robots and AI and Machine Vision Boost Agricultural and Manufacturing Capabilities, Yvonne Zhang, 2024).
Hypervisors support these AI-powered vision systems by providing a flexible platform that can manage the intense computational demands of machine learning models while ensuring that other critical tasks, like motion control and safety monitoring, are not compromised.
Simplifying System Design for the Future
By consolidating multiple functions onto a single hardware platform, hypervisors also simplify the overall design process. Instead of managing multiple controllers and interfaces, engineers can focus on optimizing software applications and machine learning models that drive the next generation of automation. This not only reduces time-to-market but also makes it easier to maintain and upgrade systems as new technologies emerge.
What are your thoughts on system consolidation? Have you considered implementing it in your next project, or do you think it’s too complex? I’d be interested to hear about your experiences and perspective on this topic.
Oh, and if you want to learn more on real-time hypervisor technology for system consolidation in mobile robots, please download our whitepaper: