Modular Platforms for Multi-Axis, Vision-Guided, and Safety-Enabled Intelligent Robotics


Robotics are the ultimate expression of industrial automation systems. Industrial control, communications, vision, machine learning, human-machine interfaces, cybersecurity and safety can be key technology considerations when building cobots, industrial robots, or other commercial applications of robotics.

Kria™ SOMs are the easiest way to build vision systems and ROS 2-based robotics using robotic-specific apps from the app store or using the Kria Robotics Stack to quickly implement and hardware accelerate your computational graph.

Along with Kria SOMs, Zynq™ UltraScale™+ MPSoCs, Versal™ adaptive SoCs, and FPGAs enable a modular approach to robotics by providing a common embedded hardware and software platform that can be reused and scaled to dramatically reduce time-to-market, form-factor, total cost of ownership while maximizing intelligence and adaptability of these assets. Specifically, adaptive SoCs and FPGAs enable:

  • Precise, deterministic control over a scalable number of axes of motion via parallel processing of control loops
  • Connectivity over diverse Industrial Ethernet standards including support for Time-Sensitive Networking (TSN)
  • Support for numerous and diverse sensor inputs, enabling sensor fusion and high-level sensors techniques
  • Real-time analytics and machine learning supporting predictive maintenance and control, remote diagnostics, digital twin, and additional edge intelligence use cases
  • Integrated human machine interface (HMI)
  • Compliance for IEC 61508 SIL 3 functional safety and IEC 62443 cybersecurity
  • Smaller physical footprint and power through highest levels of integration

Versal AI Edge series is a heterogeneous platform that provides a modular, scalable approach to robotics for all the things control, perception, navigation, self-learning, safety, and security.


Benefits of Versal AI Edge & it’s Adaptable Engines in Robotics:

  • Perception is what provides information to adapt to the environment - Sensor fusion accelerators in the adaptable engines offload the application processors
  • Adaptable engines perform low latency motor control and deterministic networking for synchronization across the system
  • Adaptable engines accelerates motion planning which is needed for navigation
  • Intelligent engines can perform AI to augment safety controls for dynamic execution, and accelerate real-time analytics with machine learning for predictive maintenance
  • Versal has built-in safety and security from bottom up for latest standards – AMD offer a simplified certification path for both secure and safe systems
Design Examples Description Device Support

Click to enlarge

Fully Integrated Robotics Control

  • Meta data processing from vision and non-vision Sensors
  • Robot Motion Planning & Safe Motion
  • Motion controller
  • Sensor bus management
  • Camera bus management, GigE Vision etc. or Ind.Ethernet like EtherCAT, Profinet, Powerlink
  • HMI
  • Safe motion safety requirements: PLe CAT3

Multi-Axis Motion Control

  • Embedded Safe Motion
  • EtherCAT Master stack (or other Industrial Ethernet) with Safety Layer
  • Safety Requirements: Ple CAT 4

Multi Axis power stage with minimum drive capabilities and safety

  • MOS FET or SiC driver stage
  • Digital Encoder
  • Safety requirements: Ple CAT 4

Machine Vision Camera

  • Sensor Interface
  • Image Signal Processing with Machine Learning
  • Network Connectivity

Kria SOM

Versal adaptive SoC

Zynq UltraScale+ MPSoC

Zynq 7000 SoC

Spartan 7 FPGA

Artix 7 FPGA

Solutions Stack

Some Industrial and Healthcare IoT products need all elements of the AMD IIoT and HcIoT Solutions Stack, all need some. The AMD IIoT and HcIoT Solutions Stack is comprised of optimized AMD and Ecosystem building blocks and solutions used across Industrial and Healthcare IoT platforms. Starting from scratch is never something you will have to do with a AMD-based Industrial or Healthcare IoT system. Minimize development time and cost and maximize design reuse on your next Industrial or Healthcare IoT platform by exploring the different elements of the AMD IIoT and HcIoT Solutions Stack.