Slide 1
Slide 1 text
SAGL: A NEW HEURISTIC FOR MULTI-
ROBOT ROUTING WITH COMPLEX
TASKS
Hong Xu*, T. K. Satish Kumar*, Dylan Johnke†,
Nora Ayanian* and Sven Koenig*
* University of Southern Calfornia, Los Angeles, CA 90089
† Cornell University, Ithaca, NY 14853
ICTAI November 7, 2016
Slide 2
Slide 2 text
AGENDA
Complex Routing Problem (CRP)
Our algorithm: SAGL
Experimental evaluation: SAGL vs others
Slide 3
Slide 3 text
AGENDA
Complex Routing Problem (CRP)
Our algorithm: SAGL
Experimental evaluation: SAGL vs others
Slide 4
Slide 4 text
MOTIVATION
Search-and-rescue: li ing heavy debris
(source: https://www.fema.gov/media-
library/assets/images/100223 )
Slide 5
Slide 5 text
COMPLEX ROUTING PROBLEM (CRP)
Multiple homogeneous robots
same moving speed
same ability to accomplish tasks
Multiple tasks in different locations
Some tasks require more than one robots to accomplish
Cooperative settings
Solution: Task visitation order for each robot
Solution evaluation: Makespan (total time required to
accomplish all tasks)
Slide 6
Slide 6 text
COMPLEX ROUTING PROBLEM EXAMPLE
Cyan: robots
Yellow: tasks requiring only 1 robot (simple tasks)
Red: tasks requiring robots ( ) (complex tasks
with a complexity level of )
Reduces to TSP-Path if only one robot and no complex
tasks.
N N ≥ 2
N
Slide 7
Slide 7 text
State-of-the-art: Approach with Reaction Functions (ARF)
[Zheng et al. '08, '11]
Based on auction mechanism
Produces good solution
Does not scale
e.g., cannot solve 20 complex tasks with a complexity
level of 2 within 1 hour
SAGL
Produces decent solution
Polynomial time complexity and scalable
Can handle high complexity levels and a large
number of complex tasks
Slide 8
Slide 8 text
AGENDA
Complex Routing Problem (CRP)
Our algorithm: SAGL
Experimental evaluation: SAGL vs others
Slide 9
Slide 9 text
EMBED COMPLEX ROUTING
PROBLEM INTO A GRAPH
A
B
C
2
2
3
We embed a problem instance into
a complete undirected edge-weighted graph
Vertices represent task and robot initial locations.
Edges represent distances between the locations.
Slide 10
Slide 10 text
ASSUMPTIONS
No collisions between robots.
Distances satisfy the triangle inequality.
Distances are symmetric.
All robots move with unit speed.
All tasks are accomplished immediately once all robots
arrive—time span required for accomplishing tasks can be
amortized into the incident edges.
Slide 11
Slide 11 text
SAGL OVERVIEW
Which robots should visit which tasks
1. Spanning tree construction
2. Task assignment
What visitation order should the robots use
3. Global visitation order determination for complex tasks
4. Local visitation order determination
Slide 12
Slide 12 text
WHICH ROBOTS SHOULD VISIT
WHICH TASKS
1. Spanning tree construction
Provides a base for task assignments
2. Task assignment
Slide 13
Slide 13 text
CONSTRUCT THE SPANNING TREE
Provides a base for task assignments:
inspired by 2-approximation TSP
Slide 14
Slide 14 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 15
Slide 15 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 16
Slide 16 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 17
Slide 17 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 18
Slide 18 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 19
Slide 19 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 20
Slide 20 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 21
Slide 21 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 22
Slide 22 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 23
Slide 23 text
CONSTRUCT THE SPANNING TREE
T
2
T
1
R
1
R
2
Slide 24
Slide 24 text
ASSIGN TASKS TO ROBOTS
According to the distances on the spanning tree.
Slide 25
Slide 25 text
WHAT VISITATION ORDER SHOULD
THE ROBOTS USE
1. Global visitation order determination for complex tasks
Prevent deadlocks
2. Local visitation order determination
Slide 26
Slide 26 text
DEADLOCKS
A B
X Y
A
B
Robot A waits at X forever.
Robot B waits at Y forever.
Slide 27
Slide 27 text
GLOBAL VISITATION ORDER OF
COMPLEX TASKS
Prevent deadlocks
Slide 28
Slide 28 text
LOCAL VISITATION ORDER
A
B
C
2
2
3
A
B
C
Path-constrained TSP [Bachrach et al. '05]:
Consistent with the global visitation order
Slide 29
Slide 29 text
AGENDA
Complex Routing Problem (CRP)
Our algorithm: SAGL
Experimental evaluation: SAGL vs others
Slide 30
Slide 30 text
EXPERIMENT SET 1
Compared with Approach with Reaction Functions (ARF)
[Zheng et al. '08, '11]
Map: 51x51 grid office environment
200 CRP instances with random vertices for
10 robots
80 simple tasks
various numbers of complex tasks with a complexity
level of 2
Slide 31
Slide 31 text
OFFICE MAP
image source: [Koenig et al. '07]
Slide 32
Slide 32 text
Percentage of instances solved by ARF within 2 minutes.
SAGL solved each of them within one second.
COMPARED WITH ARF: EFFICIENCY
Slide 33
Slide 33 text
COMPARED WITH ARF: MAKESPAN
Slide 34
Slide 34 text
EXPERIMENT SET 2
Large instances
ARF cannot solve large instances
Compared with a baseline algorithm
No spanning tree
Random global visitation order
No use of path-constrained TSP
Map: obstacle free 300x300 continuous square
15 CRP instances for random
5, 8 or 10 robots
100, 500 or 1000 tasks
max complexity levels of 2, 3 or 4
Slide 35
Slide 35 text
SAGL VS BASELINE: MAKESPAN VS NUMBER OF TASKS
Slide 36
Slide 36 text
SAGL VS BASELINE: MAKESPAN VS MAXIMUM COMPLEXITY
LEVELS
Slide 37
Slide 37 text
FUTURE WORK
Heterogenous robots
Distributed version
Flexible complexity levels (task accomplishment time
depends on number of robots)
Slide 38
Slide 38 text
CONCLUSION
SAGL is
A polynomial time solver for the Complex Routing
Problem
Four steps:
1. Spanning tree construction
2. Task assignment
3. Global visitation order determination for complex
tasks
4. Local visitation order determination
More scalable than ARF
decent solution quality