Configuration Workflow

First, specify the carrier model to be used in the simulation scene. You can select an existing carrier from an existing solution, or create a new carrier and configure its parameters.

For the specific configuration workflow of creating or editing a carrier, please refer to Scene Object Configuration.

After the carrier is determined, you need to select the target objects for simulation and configure their display color. When selecting target objects:

After the target object template is configured, click the color swatch next to Target object color to set the display color of the target object in the simulation.

Based on actual production requirements, select the placement mode for target objects, including randomly stacked and neatly arranged.

By setting a pose filter angle threshold, you can filter out pick points with excessive tilt angles. When the angle between a pick point’s Z-axis and the world coordinate system’s Z-axis exceeds this threshold, the pick point will be filtered out and will not participate in subsequent path planning.

The sorting strategy determines the priority order of picking. You can select different sorting strategies based on actual requirements. Multiple strategies are available, including descending by Z height, ascending by overlap ratio, ascending by Z-axis angle, etc. The details are as follows.

Simulation count is used to control the number of times picking is repeatedly executed. Two modes are supported: Number of consecutive picks and Number of bins.

This parameter specifies the handling method when picking fails.

First, enable Camera settings. Click Select camera, and in the drop-down list, you can select an existing camera or create a new one.

The system enables simulated point cloud noise by default to replicate the actual camera capture effect.

When enabled, you can set the point cloud filter angle to remove point clouds with angles greater than the set value, simulating the field-of-view limitation when a real camera acquires point clouds.

Click Acquire data to check the coverage of the camera’s field of view, ensuring that the capture area matches the actual requirements and that target objects are always within the camera’s field of view.

After enabling Save image data, you can set the maximum storage size and the image save path. The saved images can be used for the virtual mode in Mech-Vision.

Images are saved by default in the Data/PickSim folder under the solution folder.

After completing the above configurations, you can proceed to the picking simulation verification phase. Click Start simulation, and the system will execute picking based on the current configuration and update Pick attempts and Successful picks in real time.

After the simulation is automatically completed, one of the following prompts will appear depending on whether all target objects were picked:

To view the picking dynamics, you can switch between bins, use the slider and frame controls to flexibly view the picking effects of different bins, and play the entire picking process frame by frame.

Pick statistics is used to view the detailed results and failure reasons of previous simulations, supporting quantitative analysis of picking performance.

Click the Pick statistics button to open the statistics results window. The failure reasons can be used as a basis for optimizing picking strategies. It is recommended to pay close attention to frequently occurring failure types.

For further locating specific planning failure nodes or target objects, you can perform detailed analysis through the plan history function or by viewing logs.

In addition, by double-clicking the pick point information on a target object, you can view the vision point details. In the Vision Result Details window, click Pick point and Target object under Path planning result to view the planning status and navigate to Plan History for troubleshooting.