Our team is using the DetectingWithTensorFlowEasy sample code, and we are stuck on how to get the robot to the spike mark that the pixel/team prop is on. Would anyone have any suggestions on how we can code this without making it too complicated since we are a rookie team.?
Some hints are provided at ftc-docs here.
After exiting the main/detection loop, use the stored spike-mark location (left, center or right) to select one of three Functions (sequence of regular robot actions) to drive there and deposit the purple pixel.
To expand on what @Westside suggested, if you’re using a Pixel on a Spike Mark you have to iteratively determine which Spike Mark has the pixel on it; you must drive to the first Spike Mark and look, if there are no detections then drive to the second, look again, and if still no detections then it’s at the third Spike Mark. This is required because the Pixel has to be seen from above using the default TensorFlow model (as described here). If you’re using a custom Team Prop, it’s possible to recognize the Team Prop while looking at the Spike Marks from the starting location, which can give you an idea about which Spike Mark it’s on (based on the location of the bounding box in the camera view for the Team Prop).
Good luck!
-Danny