Home› Robotiq Products
Discussion
Back to discussions page
Tyler_Berryman
Posts: 122 Handy
How to increase the camera's field of view |
479 views
|
Answered | |
|
/ Most recent by Tyler_Berryman |
1 comment |
Tyler_Berryman
Posts: 122 Handy
Hi guys,
Since the Robotiq camera is mounted on the robot's wrist, the field of view is limited by the reach of the robot. This means that the camera's field of view is often limited to an area of about 18"x18" on a UR5. This can be problematic if you plan on using the camera to identify parts on a large table and utilizing the UR5's longer reach. However, by using multiple camera locate nodes you can increase the camera's workspace.
In this demonstration, I used 3 camera locate nodes to utilize the entire surface surrounding the UR5. These 3 camera locate nodes are looking for the exact same part. In the program, I am incrementing the variable "part" everytime time the program goes through the camera locate node. When the vision system identifies an object that matches the image template, the robot approaches the part to pick it up. If the gripper detects that there is an object in its fingertips, the robot jumps to another section of the program. This IF statement instructs the robot on where to place the part, and it removes one count from the variable "part". This is because I only want to increment the variable "part" when the vision system fails to identify a part in its field of view since I am using this variable to change snapshot positions. When the robot fails to identify a part in the 3rd and last snapshot position, the variable "part" is reset to 0 and the robot returns to the 1st snapshot position.
Here is a video of the demonstration :
https://www.youtube.com/watch?v=JyzNq4X5clI&feature=youtu.be
Since the Robotiq camera is mounted on the robot's wrist, the field of view is limited by the reach of the robot. This means that the camera's field of view is often limited to an area of about 18"x18" on a UR5. This can be problematic if you plan on using the camera to identify parts on a large table and utilizing the UR5's longer reach. However, by using multiple camera locate nodes you can increase the camera's workspace.
In this demonstration, I used 3 camera locate nodes to utilize the entire surface surrounding the UR5. These 3 camera locate nodes are looking for the exact same part. In the program, I am incrementing the variable "part" everytime time the program goes through the camera locate node. When the vision system identifies an object that matches the image template, the robot approaches the part to pick it up. If the gripper detects that there is an object in its fingertips, the robot jumps to another section of the program. This IF statement instructs the robot on where to place the part, and it removes one count from the variable "part". This is because I only want to increment the variable "part" when the vision system fails to identify a part in its field of view since I am using this variable to change snapshot positions. When the robot fails to identify a part in the 3rd and last snapshot position, the variable "part" is reset to 0 and the robot returns to the 1st snapshot position.
Here is a video of the demonstration :
https://www.youtube.com/watch?v=JyzNq4X5clI&feature=youtu.be
Tagged:
Program Robot Program 'MoveJ' If parts<2 MoveJ Waypoint_1 Gripper Move 25% Camera Locate MoveL Waypoint_2 Waypoint_3 Gripper Close Waypoint_4 Set parts=parts+1 ElseIf 2≤parts and parts≤4 MoveJ Waypoint_8 Gripper Move 25% Camera Locate MoveL Waypoint_5 Waypoint_6 Gripper Close Waypoint_7 Set parts=parts+1 ElseIf 4≤parts and parts≤6 MoveJ Waypoint_9 Gripper Move 25% Camera Locate MoveL Waypoint_11 Waypoint_10 Gripper Close Waypoint_12 Set parts=parts+1 ElseIf parts≥6 parts≔0 If rq_is_object_detected() MoveJ Approach_Tray parts≔parts-1 MoveL Waypoint_13 Gripper Move 25%Here is the program I used for the demo:Robotiq Integration Coach
[email protected]
1-418-380-2788, option 3