Inference Deployment

Inference deployment loads a trained model and connects the model output to the real robot control workflow.

Select Model

The model selection step loads the local model files required for inference:

• Select the model type and load a local checkpoint directory.
• The checkpoint directory must contain model weights, statistics, and inference configuration files.
• For VLA models, enter a task instruction to help the model understand the execution goal.
• The page checks the machine learning inference runtime plugin and provides installation or update options.

Hardware Setup

Hardware setup connects model inputs with robot outputs:

• Scan and assign camera roles to make sure the inference view source is correct.
• Configure robot serial ports, robotic arm roles, and the primary control arm.
• After confirming the hardware connection, enter the inference control page.

Inference Control

Inference control runs the model online and shows the robot execution status:

• Start or stop online inference tasks.
• View runtime status, frame rate, action queue, and prediction results in real time.
• View robot feedback data such as joint angles and gripper status.
• Use emergency stop to interrupt robot motion immediately if an abnormal situation occurs.