SLIDE 1 Robot-Assisted Discovery of Evacuation Routes in Emergency Scenarios Ettore Ferranti
Niki Trigoni Computing Laboratory, University of Oxford, UK
Multi-Service Networks, 11th July 2008
SLIDE 2 Scenario
- 1. No prior knowledge of the area’s map.
- 2. Lack of exact knowledge of agents’ and victims’ positions.
- A Group of mini robots (agents) is exploring an
area where an emergency event has just happened.
SLIDE 3
- The environment is divided into square cells.
- When an agent first moves to a cell, it deploys a
stationary sensor (tag).
- The agent then moves into one of the four adjacent cells.
Model
E N S W
SLIDE 4
Objective
Whilst exploring an unknown area, dynamically discover and maintain short evacuation routes connecting emergency exits to critical points in the area.
SLIDE 5
- Ants*
- Multiple Depth First Search+
- Brick&Mortar+
Exploration Algorithms
*J. Svennebring and S. Koenig. Building terrain-covering ant robots: A feasibility study. Auton. Robots, 2004.
+E. Ferranti, N. Trigoni and M. Levene. Brick&Mortar: An On-Line Multi-Agent Exploration Algorithm. ICRA 2007.
SLIDE 6 Ants*
*J. Svennebring and S. Koenig. Building terrain-covering ant robots: A feasibility study. Auton. Robots, 16(3):313–332, 2004.
- Each agent moves toward the least visited adjacent cell.
SLIDE 7 Ants*
*J. Svennebring and S. Koenig. Building terrain-covering ant robots: A feasibility study. Auton. Robots, 16(3):313–332, 2004.
- Each agent moves toward the least visited adjacent cell.
SLIDE 8 Multiple Depth First Search+
- Agents navigate a branch downwards to mark the
cells as .
- Agents navigate a branch upwards to mark the
cells as . Explored Visited
+E. Ferranti, N. Trigoni and M. Levene. Brick&Mortar: An On-Line Multi-Agent Exploration Algorithm. ICRA 2007.
SLIDE 9 Multiple Depth First Search+
- Agents navigate a branch downwards to mark the
cells as .
- Agents navigate a branch upwards to mark the
cells as . Explored Visited
+E. Ferranti, N. Trigoni and M. Levene. Brick&Mortar: An On-Line Multi-Agent Exploration Algorithm. ICRA 2007.
SLIDE 10 Brick&Mortar+
Agents thicken the existing walls of a room with virtual “bricks” (Visited cells).
+E. Ferranti, N. Trigoni and M. Levene. Brick&Mortar: An On-Line Multi-Agent Exploration Algorithm. ICRA 2007.
SLIDE 11 Brick&Mortar+
Agents thicken the existing walls of a room with virtual “bricks” (Visited cells).
+E. Ferranti, N. Trigoni and M. Levene. Brick&Mortar: An On-Line Multi-Agent Exploration Algorithm. ICRA 2007.
SLIDE 12 Evacuation Paths
- The distance value indicates the number of steps
from the nearest exit, which can be reached by following the pointer to the parent cell.
- Each cell has a distance value and a pointer to its
parent in the evacuation route.
Dist Dist
SLIDE 13
E E
Evacuation Path Mechanism
SLIDE 14
E E
Evacuation Path Mechanism
1 2 3 4 5 4 2 3 1 5 6 7 8 9 10 11 8 10 9
SLIDE 15
E E
Evacuation Path Mechanism
1 2 3 4 5 4 2 3 1 5 6 7 8 9 10 11 8 10 9 1 2 3 2 1 2 3 4 5
SLIDE 16
E E
Evacuation Path Mechanism
1 2 3 4 5 4 2 3 1 5 6 7 8 9 10 11 8 10 9 1 2 3 2 1 2 3 4 5
SLIDE 17
E E
Evacuation Path Mechanism
1 2 3 4 5 4 2 3 1 5 6 7 8 9 10 11 8 10 9 1 2 3 2 1 2 3 4 5 3
SLIDE 18
E E
Evacuation Path Mechanism
1 2 3 4 5 4 2 3 1 5 6 7 8 9 10 11 8 10 9 1 2 3 2 1 2 3 4 5 3 4 4
SLIDE 19
Route Discovery
E V E
SLIDE 20
Route Discovery
E V E
1 2 3 4 5 2 3 4 5 6 1 3 4 5 6 7 2 4 5 6 7 8 3 5 6 7 8 9 4 6 V 8 9 10 5 6 7 1 1 5
SLIDE 21 Route Discovery
E V E
- Brick&Mortar: 33 steps to
find a 7 cells evacuation path.
1 2 3 4 5 2 3 4 5 6 1 3 4 5 6 7 2 4 5 6 7 8 3 5 6 7 8 9 4 6 V 8 9 10 5 6 7 1 1 5
1 E 2 3 4 5 V 6
SLIDE 22 Route Discovery
E V E
- Brick&Mortar: 33 steps to
find a 7 cells evacuation path.
- Ants: 48 steps to find a 3 cells
evacuation path.
1 2 3 4 5 2 3 4 5 6 1 3 4 5 6 7 2 4 5 6 7 8 3 5 6 7 8 9 4 6 V 8 9 10 5 6 7 1 1 5 1 2 2 3 2 1
E 1 2 V
SLIDE 23 Route Discovery
E V E
- Brick&Mortar: 33 steps to
find a 7 cells evacuation path.
- MDFS: 50 steps to find a 3
cells evacuation path.
- Ants: 48 steps to find a 3 cells
evacuation path.
1 2 3 4 5 2 3 4 5 6 1 3 4 5 6 7 2 4 5 6 7 8 3 5 6 7 8 9 4 6 V 8 9 10 5 6 7 1 1 5 1 2 2 3 2 1 3 2 1 1 4
E 1 2 V
5 4
SLIDE 24
- Agent2Tag: agents communicate indirectly by
reading and updating the state of tags.
What if we change our communication assumptions?
SLIDE 25
- Agent2Tag: agents communicate indirectly by
reading and updating the state of tags.
What if we change our communication assumptions?
- Tag2Tag: tags can exchange messages to update
their state.
SLIDE 26 Evacuation Paths
- A message is sent to the adjacent neighbours
each time the distance value of a cell is changed.
(Tag2Tag)
3 3 4 5
V
2 1 1 2 3
E
9 8 5 6 10 10
SLIDE 27 Evacuation Paths
- A message is sent to the adjacent neighbours
each time the distance value of a cell is changed.
(Tag2Tag)
3 3 4 5
V
2 1 1 2 3
E
9 8 5 6 10 10 4
SLIDE 28 Evacuation Paths
- A message is sent to the adjacent neighbours
each time the distance value of a cell is changed.
(Tag2Tag)
3 3 4 5
V
2 1 1 2 3
E
9 8 5 6 10 10 4 5
SLIDE 29 Evacuation Paths
- A message is sent to the adjacent neighbours
each time the distance value of a cell is changed.
(Tag2Tag)
3 3 4 5
V
2 1 1 2 3
E
9 8 5 6 10 10 4 5 6 6
SLIDE 30 Evacuation Paths
- A message is sent to the adjacent neighbours
each time the distance value of a cell is changed.
(Tag2Tag)
3 3 4 5
V
2 1 1 2 3
E
9 8 5 6 10 10 4 5 6 6 7
SLIDE 31
Impact of Tag2Tag
SLIDE 32 B&M in Different Scenarios
B&M B&M
SLIDE 33 Conclusions
- Without Tag2Tag, faster algorithms are not better in
finding good evacuation paths. In particular, Brick&Mortar tends to be the fastest but yields longer evacuation paths.
- With Tag2Tag, all algorithms find shortest length
evacuation paths. Among them, Brick&Mortar is preferred because it is the the fastest one.
SLIDE 34
Thank you!
...questions?
