rc_reason_stack

AI-Based 3D Vision:
The Key to Your Physical AI System

rc_reason_stack: Flexible 3D Vision for Physical AI Systems

3D vision is a core capability for Physical AI systems that operate in real-world environments. The rc_reason_stack enables customers to deploy our complete AI-based 3D vision software stack directly on their own Industrial PC (IPC).

The rc_reason_stack brings proven robot perception technology to industrial edge ecosystems and provides a flexible deployment option for Physical AI and automation applications.

Run the rc_reason Software Stack on Your Own IPC

The rc_reason_stack allows the rc_reason software suite to be installed on customer-provided computing hardware. It is delivered as a set of Docker containers and is designed for deployment in industrial edge environments, including scenarios with Cyber Resilience Act (CRA) compliance requirements.

This deployment option is well suited for users who already operate IPC infrastructure or prefer to manage their own industrial computing platforms while maintaining full access to AI-based 3D vision capabilities.

Designed for the Industrial Edge

The rc_reason_stack is designed for integration into industrial robotics and automation systems that rely on reliable 3D vision for Physical AI. Its containerized architecture enables:

Seamless integration into industrial edge ecosystems

Deployment on existing IPC hardware
Scalable and maintainable software operation
Joint deployment and operation with other containers such as motion planning or quality inspection

This makes the rc_reason_stack a flexible solution for AI-based robot perception in production environments.

Same Reliable Features as the rc_cube

The rc_reason_stack offers the same core functionality as the firmware deployed on the rc_cube, including:

Support for up to four camera pipelines
Identical AI-based perception and reasoning capabilities
The same underlying software components and algorithms

The only difference is the UserSpace, which is not included by default and must be operated in a separate Docker container when required.

rc_reason_stack in a Nutshell

The rc_reason_stack is a high-performance 3D-image-processing software solution for deployment on Ubuntu machines.

Enhances computing capabilities of rc_visard and supports the rc_viscore, the Basler Stereo ace, the Orbbec and the Zivid camera
Delivers real-time camera images and depth images, which can be used to compute 3D point clouds
Provides confidence and error images as quality measures for each image acquisition
Intuitive WebGUI (user interface) and standardized interfaces make it compatible with all major image processing libraries

With optionally available software modules the rc_reason_stack provides out-of-the-box solutions for object detection and robotic pick-and-place applications.

Additional Information about the rc_reason_stack

Components' Minimum Versions

Ubuntu	24.04 LTS
NVIDIA GPU	Any T`RTX with minimum 8GB VRAM, or better. (Tested with Nvidia RTX A4000, RTX 4000 Ada, RTX 3080, RTX 4070, RTX 4080)
Docker	20.10+
NVIDIA Driver	535+ (the guide uses nvidia-driver-575-server)

Hardware & Interfaces

Pipelines	supports four camera pipelines
Supported Camera Types	rc_visard, rc_visard NG, rc_viscore, zivid, Orbbec, Stero ace.
Interfaces	WebGUI, REST-API, Generic Robot Interface (GRI), OPC UA, KUKA Ethernet KRL, gRCP image stream

Documentation & Download

For more detailed information about our products and our software tools for your robot vision solution, please visit our Knowledge Base:

Contact Us

Free Feasibility & Demo

You would like to find out whether our portfolio is suitable for your robotic application? Simply request a feasibility study free of charge, and get a live demo of our products.

Try & Buy

Would you like to try out one of our sensors and software solutions? Our Try-&-Buy-option gives you the chance to test our products before you decide, and to be sure you make the right choice for your application.