The inside of a tire is manually inspected for surface defects and pentrations by foreign objects. It can be physically awkward and difficult to see small objects manually so the vision system handles that task.
For this pilot project, a camera on a linear slide and rotating mirror constrained to a 12” x 3” package was mounted on the end of linear arm and joysticked into the center of the tire. Laser gauges provide feedback to the operator for its ideal placement. Interior is laser profiled to calculate optimal camera distance such that the entire field of view is within focus. The mirror is centered to the field of view and the tire is rotated while the camera acquires a series of images. After each series of images, the mirror is adjusted to the next field of view and the process is repeated until the interior is fully imaged. Images are then presented to the operator on a 34” curve monitor for manual inspection. The operator reviews the images and draws boxes around possible defects. Afterwards this information is archived with an identifier. At a later date, this data can be used to train a Deep Learning system.
A scheduling system dictates the components to be assembled on a diesel engine head. This information is written to an RFID tag that is attached to a carriage. The client needed a way to validate the correct engine head was loaded onto the carriage and get its information (production date, parent, and serial number) married to the RFID tag. (7) different head types will need to be handled by this system.
Multiple Cognex Dataman cameras are oriented in a work cell. As the carriage enters into the work cell and stops, an RFID reader gets the part type and passes that to the HMI software, which triggers 1 or 2 cameras. The cameras extract the date, parent, and serial number for a 2D matrix dot pin barcode and the HMI software writes the data to the RFID tag and an edge database. If the parent number is not valid for the current part type, the error is flagged for an operator to review. In the event of a poor barcode, the operator has the ability to enter the information manually.