Palletizing

RoboDK Workstation

2MB

pallet_workstation.rdk

Robot Program File - Simulation

# pallet setup
rows = 2
columns = 4
layers = 8

LEFT_PALLET = "LEFT PALLET"
RIGHT_PALLET = "RIGHT PALLET"
selected_pallet = LEFT_PALLET

LEFT_PALLET_FEATURE = p[-0.580, 0.528, -0.860, 0, 0, 0]
RIGHT_PALLET_FEATURE = p[0.580, -0.528, -0.860, 0, 0, 3.1415]
pallet_feature = LEFT_PALLET_FEATURE

# box setup
box_length = 300
box_width = 400
box_height = 200

PICK_FEATURE = p[-0.860, -0.211, -0.350, 0, 0, 1.571]

# start from setup
FIRST_BOX = "FIRST BOX"
CURRENT_BOX = "CURRENT BOX"
BOX_AT = "BOX AT"
start_from = FIRST_BOX

start_row = 0
start_column = 0
start_layer = 0

# pallet state
current_row = 0
current_column = 0
current_layer = 0


def get_pick_pose():
    x = box_length * 0.0005
    y = box_width * 0.0005
    z = box_height * 0.001
    return pose_trans(PICK_FEATURE, p[x, y, z, 0, 3.1415, 0])
end

def get_place_pose():
    x = box_length * 0.0005 + current_column * box_length * 0.001
    y = box_width * 0.0005 + current_row * box_width * 0.001
    z = box_height * 0.001 + current_layer * box_height * 0.001
    return pose_trans(pallet_feature, p[x, y, z, 0, 3.1415, 0])
end

def get_approach_pose(pose, offset=[0, 0, 0]):
    return pose_add(pose, p[offset[0], offset[1], offset[2], 0, 0, 0])
end

def grasp():
    sim_grasp("box_{}_{}_{}".format(current_row, current_column, current_layer))
end

def release():
    sim_release()
end



def start_task():

    # start from condition
    if start_from == FIRST_BOX:
        current_row = 0
        current_column = 0
        current_layer = 0
    elif start_from == BOX_AT:
        current_row = start_row - 1
        current_column = start_column - 1
        current_layer = start_layer - 1
    else:
        current_row = current_row
        current_column = current_column
        current_layer = current_layer
    end

    # selected pallet
    if selected_pallet == LEFT_PALLET:
        pallet_feature = LEFT_PALLET_FEATURE
    else:
        pallet_feature = RIGHT_PALLET_FEATURE
    end

    while current_layer < layers:
        while current_column < columns:
            while current_row < rows:
                # pick action
                pick_pose = get_pick_pose()
                approach_pose = get_approach_pose(pick_pose, offset=[0, 0, 0.1])
                current_pose = get_actual_tcp_pose()
                if approach_pose[2] < current_pose[2] + 0.05:
                    approach_pose[2] = current_pose[2] + 0.05
                end
                movej(approach_pose)
                movel(pick_pose)
                grasp()

                place_pose = get_place_pose()
                place_approach_pose = get_approach_pose(place_pose, offset=[0.1, 0.1, box_height * 0.001 + 0.05])

                if place_approach_pose[2] > PICK_FEATURE[2] +  box_height * 0.001 + 0.05:
                    approach_pose[2] = place_approach_pose[2]
                end

                movel(approach_pose)

                # place action
                place_pose = get_place_pose()
                approach_pose = get_approach_pose(place_pose, offset=[0.1, 0.1, box_height * 0.001 + 0.05])
                movej(approach_pose)
                movel(place_pose)
                release()
                approach_pose = get_approach_pose(place_pose, offset=[0, 0, 0.1])
                if approach_pose[2] < PICK_FEATURE[2] +  box_height * 0.001 + 0.05:
                    approach_pose[2] = PICK_FEATURE[2] +  box_height * 0.001 + 0.05
                end
                movel(approach_pose)

                current_row = current_row + 1
            end
            current_row = 0
            current_column = current_column + 1
        end
        current_row = 0
        current_column = 0
        current_layer = current_layer + 1
    end
end

start_task()