Precalculated pathing

import os
from vidigi.utils import EventPosition, create_event_position_df
from vidigi.prep import reshape_for_animations, generate_animation_df
from vidigi.animation import generate_animation, animate_activity_log
import pandas as pd
import plotly.io as pio
pio.renderers.default = "notebook"

View Imported Code, which has had logging steps added at the appropriate points in the ‘model’ class

import simpy
import random
from typing import Tuple
from vidigi.logging import EventLogger


# ---------------------------
# Warehouse Layout (scaled)
# ---------------------------
# Coordinates in pixels for animation space
LAYOUT = {
    # Packing & maintenance
    "packing": (270, 270),
    "maintenance": (600, 300),

    # 8 pickup points
    "pickup_1": (25, 500),
    "pickup_2": (180, 500),
    "pickup_3": (70, 500),
    "pickup_4": (550, 500),
    "pickup_5": (25, 100),
    "pickup_6": (180, 100),
    "pickup_7": (370, 100),
    "pickup_8": (550, 100),
}

PICKUP_POINTS = list(name for name in LAYOUT if name.startswith("pickup"))

SPEED = 100.0  # pixels per minute
PACKING_TIME = 5
PACKAGES_PER_BATCH = (1, 5)
OTHER_TASK_TIME = (2, 4)
OTHER_TASK_PROB = 0.3
SIM_DURATION = 60*24


def travel_time(pos_a: Tuple[int, int], pos_b: Tuple[int, int]) -> float:
    """Calculate travel time between two coordinates."""
    dist = ((pos_a[0] - pos_b[0]) ** 2 + (pos_a[1] - pos_b[1]) ** 2) ** 0.5
    return dist / SPEED


class PackingRobot:
    def __init__(self, env, name, logger):
        self.env = env
        self.name = name
        self.logger = logger
        self.pos = LAYOUT["packing"]  # start at packing station
        self.env.process(self.poll_position())
        self.logger.log_arrival(entity_id=self.name,
                                x=self.pos[0], y=self.pos[1])
        self.logger.log_queue(entity_id=self.name, event="packing",
                               x=self.pos[0], y=self.pos[1])

    def poll_position(self):
        """Logs position every 1 sim time unit, even if idle."""
        while True:
            self.logger.log_custom_event(entity_id=self.name,
                                         event_type="position_poll",
                                         event="position",
                                         x=self.pos[0], y=self.pos[1])
            yield self.env.timeout(1)

    def move_to(self, location_name, pathway, outbound=True):
        """Move robot to a location using Manhattan path.
        outbound=True: horizontal then vertical
        outbound=False: retrace return path (vertical then horizontal)
        """
        destination = LAYOUT[location_name]
        start_x, start_y = self.pos
        dest_x, dest_y = destination

        if outbound:
            sequence = [("x", dest_x - start_x), ("y", dest_y - start_y)]
        else:
            sequence = [("y", dest_y - start_y), ("x", dest_x - start_x)]

        for axis, delta in sequence:
            if delta != 0:
                travel_time = abs(delta) / SPEED
                steps = int(travel_time)
                remaining = travel_time - steps
                move_per_unit = delta / travel_time

                for _ in range(steps):
                    yield self.env.timeout(1)
                    if axis == "x":
                        self.pos = (self.pos[0] + move_per_unit, self.pos[1])
                    else:
                        self.pos = (self.pos[0], self.pos[1] + move_per_unit)
                if remaining > 0:
                    yield self.env.timeout(remaining)
                    if axis == "x":
                        self.pos = (self.pos[0] + move_per_unit * remaining, self.pos[1])
                    else:
                        self.pos = (self.pos[0], self.pos[1] + move_per_unit * remaining)


    def pickup_packages(self, count, pickup_name):
        yield self.env.process(self.move_to(pickup_name, "to_pickup", outbound=True))
        # self.logger.log_custom_event(entity_id=self.name, event_type="action",
        #                              event=f"picked_up_{count}_packages",
        #                              x=self.pos[0], y=self.pos[1],
        #                              location=pickup_name)

        yield self.env.process(self.move_to("packing", "to_packing", outbound=False))


        self.logger.log_queue(entity_id=self.name,
                                    event=pickup_name,
                                    x=self.pos[0], y=self.pos[1],
                                    package_count=count)
        for i in range(count):
            yield self.env.timeout(PACKING_TIME)

        if random.random() < OTHER_TASK_PROB:
            yield self.env.process(self.other_task())

        # Go back to the packing station
        self.logger.log_queue(entity_id=self.name,
                                    event="packing",
                                    x=self.pos[0], y=self.pos[1])



    def other_task(self):
        yield self.env.process(self.move_to("maintenance", "to_maintenance"))
        task_time = random.randint(*OTHER_TASK_TIME)
        self.logger.log_queue(entity_id=self.name,
                                     event="maintenance",
                                     x=self.pos[0], y=self.pos[1],
                                     task_duration_mins=task_time
                                     )
        yield self.env.timeout(task_time)
        yield self.env.process(self.move_to("packing", "return_from_maintenance"))
        # Logging of return to packing location will be handled in pickup_packages process


def package_arrival(env, robot):
    while True:
        yield env.timeout(random.randint(4, 8))
        num_packages = random.randint(*PACKAGES_PER_BATCH)
        pickup_name = random.choice(PICKUP_POINTS)
        yield env.process(robot.pickup_packages(num_packages, pickup_name))




# ---------------------------
# Running the simulation
# ---------------------------
if __name__ == "__main__":
    env = simpy.Environment()
    logger = EventLogger(env=env)
    robot = PackingRobot(env, "RoboPack-1", logger)
    env.process(package_arrival(env, robot))
    env.run(until=SIM_DURATION)
    logger.log_departure(entity_id=robot.name,
                            x=robot.pos[0], y=robot.pos[1])

    logger.to_csv("robot_log.csv")

# Define positions for animation
event_positions = create_event_position_df([
    EventPosition(event='arrival', x=0, y=550, label="Entrance"),

    EventPosition(event='pickup_1', x=40, y=500, label="Pickup 1"),
    EventPosition(event='pickup_2', x=170, y=500, label="Pickup 2"),
    EventPosition(event='pickup_3', x=350, y=500, label="Pickup 3"),
    EventPosition(event='pickup_4', x=500, y=500, label="Pickup 4"),
    EventPosition(event='pickup_5', x=40, y=60, label="Pickup 5"),
    EventPosition(event='pickup_6', x=170, y=60, label="Pickup 6"),
    EventPosition(event='pickup_7', x=350, y=60, label="Pickup 7"),
    EventPosition(event='pickup_8', x=500, y=60, label="Pickup 8"),

    EventPosition(event='packing', x=300, y=300, label="Packing"),
    EventPosition(event='maintenance', x=600, y=300, label="Maintenance"),

    EventPosition(event='depart', x=650, y=50, label="Exit")
])

event_log_df = pd.read_csv("robot_log.csv")
event_log_df.head()

	entity_id	event_type	event	time	pathway	run_number	timestamp	resource_id	x	y	package_count	task_duration_mins
0	RoboPack-1	arrival_departure	arrival	0.0	NaN	NaN	NaN	NaN	270.0	270.0	NaN	NaN
1	RoboPack-1	queue	packing	0.0	NaN	NaN	NaN	NaN	270.0	270.0	NaN	NaN
2	RoboPack-1	position_poll	position	0.0	NaN	NaN	NaN	NaN	270.0	270.0	NaN	NaN
3	RoboPack-1	position_poll	position	1.0	NaN	NaN	NaN	NaN	270.0	270.0	NaN	NaN
4	RoboPack-1	position_poll	position	2.0	NaN	NaN	NaN	NaN	270.0	270.0	NaN	NaN

STEP_SNAPSHOT_MAX = 999
LIMIT_DURATION = int(max(event_log_df[event_log_df["event_type"]!="position_poll"]['time']))
WRAP_QUEUES_AT = 999

event_log_df_filtered = event_log_df[~event_log_df["event_type"].isin(["position_poll", "action"])][['entity_id', 'event_type', 'event', 'time', 'pathway']]
event_log_df_filtered.head(10)

	entity_id	event_type	event	time	pathway
0	RoboPack-1	arrival_departure	arrival	0.0	NaN
1	RoboPack-1	queue	packing	0.0	NaN
13	RoboPack-1	queue	pickup_2	10.4	NaN
23	RoboPack-1	queue	maintenance	19.0	NaN
29	RoboPack-1	queue	packing	24.6	NaN
46	RoboPack-1	queue	pickup_8	40.6	NaN
67	RoboPack-1	queue	packing	60.6	NaN
86	RoboPack-1	queue	pickup_1	78.1	NaN
97	RoboPack-1	queue	packing	88.1	NaN
114	RoboPack-1	queue	pickup_8	104.1	NaN

We’ll first run this while allowing vidigi to handle the pathing. This means that the path between each step will be interpolated.

animate_activity_log(
    event_log=event_log_df_filtered,
    event_position_df=event_positions,
    wrap_queues_at=WRAP_QUEUES_AT,
    limit_duration=LIMIT_DURATION,
    step_snapshot_max=STEP_SNAPSHOT_MAX,
    every_x_time_units=1,
    debug_mode=True,
    custom_entity_icon_list=["🤖"]
)

Animation function called at 16:38:11
Iteration through time-unit-by-time-unit logs complete 16:38:17
Snapshot df concatenation complete at 16:38:17
Reshaped animation dataframe finished construction at 16:38:17
Placement dataframe finished construction at 16:38:17
Output animation generation complete at 16:38:22
Total Time Elapsed: 10.60 seconds

If we were to put in a background illustrating the paths the robots should follow, this will look bad:

animate_activity_log(
    event_log=event_log_df_filtered,
    event_position_df=event_positions,
    wrap_queues_at=WRAP_QUEUES_AT,
    limit_duration=LIMIT_DURATION,
    step_snapshot_max=STEP_SNAPSHOT_MAX,
    every_x_time_units=1,
    debug_mode=True,
    display_stage_labels=False,
    custom_entity_icon_list=["🤖"],
    add_background_image="https://raw.githubusercontent.com/hsma-tools/vidigi/refs/heads/main/examples/example_16_packing_robot/warehouse.png",
    background_image_opacity=1, # New parameter in 1.1.0
    override_x_max=650,
    override_y_max=550,
    plotly_width=1300,
    plotly_height=800,
)

Animation function called at 16:38:22
Iteration through time-unit-by-time-unit logs complete 16:38:28
Snapshot df concatenation complete at 16:38:28
Reshaped animation dataframe finished construction at 16:38:28
Placement dataframe finished construction at 16:38:28
Output animation generation complete at 16:38:32
Total Time Elapsed: 10.22 seconds

However, in this particular model, we have been polling the location of the robot after every step.

This might be imporant to visualise accurately in certain models - for example, to demonstrate why a robot might have to wait for another robot to move out of the way, and to assure stakeholders that such a thing has been recorded accurately.

Let’s see how we can combine vidigi with these polled locations.

full_entity_df = reshape_for_animations(
    event_log=event_log_df[~event_log_df["event_type"].isin(["position_poll", "action"])][['entity_id', 'event_type', 'event', 'time', 'pathway']],
    limit_duration=LIMIT_DURATION,
    step_snapshot_max=STEP_SNAPSHOT_MAX,
    every_x_time_units=1,
    debug_mode=True
    )

full_entity_df

Iteration through time-unit-by-time-unit logs complete 16:38:38
Snapshot df concatenation complete at 16:38:38

	index	entity_id	event_type	event	time	pathway	rank	snapshot_time
0	1	RoboPack-1	queue	packing	0.0	NaN	1.0	0
1	1	RoboPack-1	queue	packing	0.0	NaN	1.0	1
2	1	RoboPack-1	queue	packing	0.0	NaN	1.0	2
3	1	RoboPack-1	queue	packing	0.0	NaN	1.0	3
4	1	RoboPack-1	queue	packing	0.0	NaN	1.0	4
...	...	...	...	...	...	...	...	...
1436	1535	RoboPack-1	queue	packing	1428.4	NaN	1.0	1436
1437	1535	RoboPack-1	queue	packing	1428.4	NaN	1.0	1437
1438	1535	RoboPack-1	queue	packing	1428.4	NaN	1.0	1438
1439	1535	RoboPack-1	queue	packing	1428.4	NaN	1.0	1439
1440	1547	RoboPack-1	arrival_departure	depart	1440.0	NaN	1.0	1440

1441 rows × 8 columns

full_entity_df_plus_pos = generate_animation_df(
    full_entity_df=full_entity_df,
    event_position_df=event_positions,
    wrap_queues_at=WRAP_QUEUES_AT,
    step_snapshot_max=STEP_SNAPSHOT_MAX,
    debug_mode=True,
    custom_entity_icon_list=["🤖"]
    )

full_entity_df_plus_pos

Placement dataframe finished construction at 16:38:38

	index	entity_id	event_type	event	time	pathway	rank	snapshot_time	x	y_final	label	resource	x_final	row	y	icon	opacity
0	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	19	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
1	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	20	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
2	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	21	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
3	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	22	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
4	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	23	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1435	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1385	500	60.0	Pickup 8	None	500.0	0.0	NaN	🤖	1.0
1436	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1386	500	60.0	Pickup 8	None	500.0	0.0	NaN	🤖	1.0
1437	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1387	500	60.0	Pickup 8	None	500.0	0.0	NaN	🤖	1.0
1438	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1388	500	60.0	Pickup 8	None	500.0	0.0	NaN	🤖	1.0
1439	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1389	500	60.0	Pickup 8	None	500.0	0.0	NaN	🤖	1.0

1440 rows × 17 columns

Now we can replace our calculated x_final and y_final coordinates with the values from the polling.

First, we pull this data out of our event logs and rename the columns to match the relevant columns in our transformed event logging dataset.

entity_position_df = event_log_df[event_log_df["event_type"]=="position_poll"][['entity_id', 'time', 'x', 'y']].reset_index(drop=True)
entity_position_df = entity_position_df.rename(columns={"x": "x_final", "y": "y_final", "time": "snapshot_time"})
entity_position_df

	entity_id	snapshot_time	x_final	y_final
0	RoboPack-1	0.0	270.0	270.0
1	RoboPack-1	1.0	270.0	270.0
2	RoboPack-1	2.0	270.0	270.0
3	RoboPack-1	3.0	270.0	270.0
4	RoboPack-1	4.0	270.0	270.0
...	...	...	...	...
1435	RoboPack-1	1435.0	25.0	270.0
1436	RoboPack-1	1436.0	25.0	370.0
1437	RoboPack-1	1437.0	25.0	470.0
1438	RoboPack-1	1438.0	25.0	500.0
1439	RoboPack-1	1439.0	25.0	400.0

1440 rows × 4 columns

We then drop our original x_final and y_final columns from our transformed dataset, replacing them with those from the polling in our model.

full_entity_df_plus_pos_manual_locations = (
    full_entity_df_plus_pos
    .drop(columns=["x_final", "y_final"])
    .merge(entity_position_df, on=["entity_id", "snapshot_time"])
)

full_entity_df_plus_pos_manual_locations

	index	entity_id	event_type	event	time	pathway	rank	snapshot_time	x	label	resource	row	y	icon	opacity	x_final	y_final
0	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	19	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	270.0
1	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	20	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
2	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	21	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
3	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	22	600	Maintenance	None	0.0	NaN	🤖	1.0	500.0	300.0
4	23	RoboPack-1	queue	maintenance	19.0	NaN	1.0	23	600	Maintenance	None	0.0	NaN	🤖	1.0	400.0	300.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1435	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1385	500	Pickup 8	None	0.0	NaN	🤖	1.0	270.0	270.0
1436	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1386	500	Pickup 8	None	0.0	NaN	🤖	1.0	270.0	270.0
1437	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1387	500	Pickup 8	None	0.0	NaN	🤖	1.0	370.0	270.0
1438	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1388	500	Pickup 8	None	0.0	NaN	🤖	1.0	470.0	270.0
1439	1472	RoboPack-1	queue	pickup_8	1370.7	NaN	1.0	1389	500	Pickup 8	None	0.0	NaN	🤖	1.0	600.0	270.0

1440 rows × 17 columns

Finally, we then generate our animation as usual - making sure to refer to our updated dataframe.

fig = generate_animation(
        full_entity_df_plus_pos=full_entity_df_plus_pos_manual_locations.sort_values(['entity_id', 'snapshot_time']),
        event_position_df= event_positions,
        simulation_time_unit="seconds",
        display_stage_labels=False,
        setup_mode=False,
        start_time="07:00:00",
        time_display_units="%H:%M:%S",
        debug_mode=True,
        add_background_image="https://raw.githubusercontent.com/hsma-tools/vidigi/refs/heads/main/examples/example_16_packing_robot/warehouse.png",
        background_image_opacity=1, # New parameter in 1.1.0
        override_x_max=650,
        override_y_max=550,
        plotly_width=1300,
        plotly_height=800,
        entity_icon_size=50,
        frame_duration=200,
        frame_transition_duration=300

    )

fig

Output animation generation complete at 16:38:43

Let’s finally repeat this with an instance with multiple packers.

In this very simplistic model, blocking of corridors is not implemented - but is the kind of thing that would visualise well with this kind of pre-calculated movement.

event_log_df_multiple = pd.read_csv("robot_log_multiple.csv")

full_entity_df_multiple = reshape_for_animations(
    event_log=event_log_df_multiple[~event_log_df_multiple["event_type"].isin(["position_poll", "action"])][['entity_id', 'event_type', 'event', 'time', 'pathway']],
    limit_duration=LIMIT_DURATION,
    step_snapshot_max=STEP_SNAPSHOT_MAX,
    every_x_time_units=1,
    debug_mode=True
    )

full_entity_df_plus_pos_multiple = generate_animation_df(
    full_entity_df=full_entity_df_multiple,
    event_position_df=event_positions,
    wrap_queues_at=WRAP_QUEUES_AT,
    step_snapshot_max=STEP_SNAPSHOT_MAX,
    debug_mode=True,
    custom_entity_icon_list=["🤖"]
    # custom_entity_icon_list=["🟦", "🟫", "🟪", "🟧", "🟥", "🟨", "🟩", "◻️"]
    )

full_entity_df_plus_pos_multiple

Iteration through time-unit-by-time-unit logs complete 16:38:51
Snapshot df concatenation complete at 16:38:51
Placement dataframe finished construction at 16:38:51

	index	entity_id	event_type	event	time	pathway	rank	snapshot_time	x	y_final	label	resource	x_final	row	y	icon	opacity
0	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	25	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
1	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	26	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
2	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	27	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
3	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	28	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
4	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	29	600	300.0	Maintenance	None	600.0	0.0	NaN	🤖	1.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
11515	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	597	350	60.0	Pickup 7	None	350.0	0.0	NaN	🤖	1.0
11516	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	598	350	60.0	Pickup 7	None	350.0	0.0	NaN	🤖	1.0
11517	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	599	350	60.0	Pickup 7	None	350.0	0.0	NaN	🤖	1.0
11518	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	600	350	60.0	Pickup 7	None	350.0	0.0	NaN	🤖	1.0
11519	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	601	350	60.0	Pickup 7	None	350.0	0.0	NaN	🤖	1.0

11520 rows × 17 columns

entity_position_df_multiple = (
    event_log_df_multiple[event_log_df_multiple["event_type"]=="position_poll"]
    [['entity_id', 'time', 'x', 'y']]
    .reset_index(drop=True)
    .rename(columns={"x": "x_final", "y": "y_final", "time": "snapshot_time"})
    )

# entity_position_df_multiple["snapshot_time"] = entity_position_df_multiple["snapshot_time"].astype('int')

entity_position_df_multiple

	entity_id	snapshot_time	x_final	y_final
0	RoboPack-1	0.0	270.0	270.0
1	RoboPack-2	0.0	285.0	270.0
2	RoboPack-3	0.0	300.0	270.0
3	RoboPack-4	0.0	315.0	270.0
4	RoboPack-5	0.0	270.0	300.0
...	...	...	...	...
11515	RoboPack-4	1439.0	300.0	300.0
11516	RoboPack-5	1439.0	270.0	300.0
11517	RoboPack-6	1439.0	300.0	300.0
11518	RoboPack-7	1439.0	300.0	300.0
11519	RoboPack-8	1439.0	315.0	300.0

11520 rows × 4 columns

full_entity_df_plus_pos_manual_locations_multiple = (
    full_entity_df_plus_pos_multiple
    .drop(columns=["x_final", "y_final"])
    .merge(entity_position_df_multiple, on=["entity_id", "snapshot_time"])
)

full_entity_df_plus_pos_manual_locations_multiple

	index	entity_id	event_type	event	time	pathway	rank	snapshot_time	x	label	resource	row	y	icon	opacity	x_final	y_final
0	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	25	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
1	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	26	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
2	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	27	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
3	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	28	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
4	224	RoboPack-8	queue	maintenance	24.9	NaN	1.0	29	600	Maintenance	None	0.0	NaN	🤖	1.0	600.0	300.0
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
11515	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	597	350	Pickup 7	None	0.0	NaN	🤖	1.0	270.0	270.0
11516	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	598	350	Pickup 7	None	0.0	NaN	🤖	1.0	270.0	270.0
11517	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	599	350	Pickup 7	None	0.0	NaN	🤖	1.0	270.0	270.0
11518	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	600	350	Pickup 7	None	0.0	NaN	🤖	1.0	270.0	270.0
11519	4934	RoboPack-1	queue	pickup_7	576.6	NaN	1.0	601	350	Pickup 7	None	0.0	NaN	🤖	1.0	270.0	270.0

11520 rows × 17 columns

fig_multiple = generate_animation(
        full_entity_df_plus_pos=full_entity_df_plus_pos_manual_locations_multiple.sort_values(['entity_id', 'snapshot_time']),
        event_position_df= event_positions,
        simulation_time_unit="seconds",
        display_stage_labels=False,
        setup_mode=False,
        start_time="07:00:00",
        time_display_units="%H:%M:%S",
        debug_mode=True,
        add_background_image="https://raw.githubusercontent.com/hsma-tools/vidigi/refs/heads/main/examples/example_16_packing_robot/warehouse.png",
        background_image_opacity=1, # New parameter in 1.1.0
        override_x_max=650,
        override_y_max=550,
        plotly_width=1300,
        plotly_height=800,
        entity_icon_size=30

    )

fig_multiple

Output animation generation complete at 16:38:56