One kit. Teach, train, deploy.
SigmaKit is delivered as a complete package — AI software with robotic hands, teaching devices, and sensor modules — so it can be used easily without deep expertise in robotics or AI.
SigmaKit is delivered as a complete package — AI software with robotic hands, teaching devices, and sensor modules — so it can be used easily without deep expertise in robotics or AI.
SigmaKit combines AI software with robotic hands, teaching devices, and sensor modules. To ensure intuitive application across diverse manufacturing environments, our robotic hands and teaching devices will continue to expand in multiple form factors.
Designed for simplicity, SigmaKit requires no complex coding, AI expertise, or special equipment — making it accessible for anyone on the shop floor.
Teach tasks intuitively and quickly — capturing operator know-how with minimal effort to maximize training efficiency.
Powered by advanced AI learning, SigmaKit remains resilient to variability, delivering consistent performance even in challenging High-Mix, High-Variance environments.
SigmaKit also supports Hand-Held Teaching, enabling task demonstrations without direct robot manipulation. Using a glove-based interface, operators can capture high-fidelity motion and interaction data independently of robot hardware.
This decouples data acquisition from deployment, accelerating task learning and system scalability.
The hand is the data interface.
With our intuitive data collection tools, simply demonstrate the target task repeatedly. Each teleoperated motion is captured as image and motion data, quickly building a training dataset. Depending on the task, a model can be generated in less than a day.
SigmaKit makes training efficient by allowing you to easily filter, review, and edit collected data. Experience the convenience and distinct advantages built into SigmaKit's workflow.
From accumulated image and motion data, CarbonSix's AI learns and produces a robotic motion intelligence (“Skill”) that enables the robot to execute complex tasks.
The trained intelligence can be deployed directly to unstructured tasks. Even when scaling across production lines or additional sites, rollout requires minimal adjustments, enabling seamless mass deployment.
