With MIRAI 2, it is even easier to enable robots to reliably perform automation tasks with variance and deploy the system across robots fleets.
Unlike standard computer-vision systems,
the MIRAI system:
Analyze the task and the sources of variance that MIRAI will be trained to handle in that task. Equip your robot with a suitable 2D camera, an end-effector, and other needed hardware.
Familiarize the system with the movement you want it to perform and the variance it may come across. As you do, a camera records the scene.
We run the image data through an AI algorithm to prepare a mathematical model, called a “skill.” A skill tells the robot how to behave while performing an action, no matter the situation.
The MIRAI-guided robot can now perform your action and handle changing situations. You use the MIRAI skill as part of an overall task, like picking or insertion. Skills are tested and refined through further training.
Share skills with multiple robots, at the same site or elsewhere.
Robots guided by MIRAI can react to changing factory conditions. The MIRAI software consists of robot-control code on an edge device, a cloud-based machine learning service, and an app for training the system.
A customer-success team will work with you to set up MIRAI in your factory, deploy the automation solution, and keep it live 24/7. We can also help you to procure any needed hardware for the integration, including suitable 2D cameras and edge devices.
Using MIRAI, we don’t have the need for experienced programmers. We have even managed to optimize cycle times by 20% on our own. That's exactly the flexibility we need.