Content uploaded by Kuppan Chetty Ramanathan
Author content
All content in this area was uploaded by Kuppan Chetty Ramanathan on Oct 20, 2020
Content may be subject to copyright.
Content uploaded by Kuppan Chetty Ramanathan
Author content
All content in this area was uploaded by Kuppan Chetty Ramanathan on Sep 10, 2020
Content may be subject to copyright.
Content uploaded by Seenu Saravanan
Author content
All content in this area was uploaded by Seenu Saravanan on Oct 04, 2020
Content may be subject to copyright.
Review on state-of-the-art dynamic task
allocation strategies for
multiple-robot systems
Seenu N., Kuppan Chetty R.M. and Ramya M.M.
Centre for Automation and Robotics, School of Mechanical Sciences, Hindustan Institute of Technology and Science, Chennai, India, and
Mukund Nilakantan Janardhanan
School of Engineering, University of Leicester, Leicester, UK
Abstract
Purpose –This paper aims to present a concise review on the variant state-of-the-art dynamic task allocation strategies. It presents a thorough
discussion about the existing dynamic task allocation strategies mainly with respect to the problem application, constraints, objective functions and
uncertainty handling methods.
Design/methodology/approach –This paper briefs the introduction of multi-robot dynamic task allocation problem and discloses the challenges
that exist in real-world dynamic task allocation problems. Numerous task allocation strategies are discussed in this paper, and it establishes the
characteristics features between them in a qualitative manner. This paper also exhibits the existing research gaps and conducive future research
directions in dynamic task allocation for multiple mobile robot systems.
Findings –This paper concerns the objective functions, robustness, task allocation time, completion time, and task reallocation feature for
performance analysis of different task allocation strategies. It prescribes suitable real-world applications for variant task allocation strategies and
identifies the challenges to be resolved in multi-robot task allocation strategies.
Originality/value –This paper provides a comprehensive review of dynamic task allocation strategies and incites the salient research directions to
the researchers in multi-robot dynamic task allocation problems. This paper aims to summarize the latest approaches in the application of
exploration problems.
Keywords Multiple mobile robots, Dynamic task allocation, Market-based task allocation, Behavior-based task allocation, Task clustering,
Heuristic task allocation
Paper type General review
1. Introduction
A cooperative multiple-robot system is one of the most
extensive research domains in robotics (Fang et al., 2018;
Palmer et al.,2018). A multiple-robot system (MRS) deploys a
number of cooperative robots in a coordinated fashion to
execute and accomplish tasks (D’Emidio and Khan, 2017).
Deployment of a single robot to perform such complicated
tasks is time-consuming and exhausting (Li and Li, 2017).
Whereas, deployment of multiple robots overwhelms the
drawbacks of a single-robot system, are sufficient to perform
complex tasks faster than a single robot in a distributed
manner. It provides the flexibility to manipulate the robots
failure, self-reconfiguration, high fault tolerance and robustness
(Palmer et al.,2018). The applications of an MRS are in
numerous fields such as manufacturing, construction, mining,
inspection (Liu et al.,2017), warehouses (Tsang et al.,2018),
surveillance (Farinelli et al.,2017), defence applications (Jha
and Nair, 2017) agriculture, exploration of underwater, space
and land, search and rescue operations (Rishwaraj and
Ponnambalam, 2017).
In the paradigm of the multi-robot system, there are two
coordination methodologies: centralized and distributed
(Johnson et al.,2016;Semwal et al., 2017). In centralized
coordination, a server monitors the essential parameters such as
relative position, status of the task, battery capacity of
individual robots in the team. The server also identifies the
most competent robot to execute a task. This system relies on a
central server for successful task allocation; however, it
becomes futile when the central server fails (Li et al., 2017a).
Therefore, this coordination is befitting to a small team of
robots with rigidly connected networks (Liu et al., 2017). In
distributed coordination, individual robots allocate their own
tasks. It requires neither a globally connected network nor a
central server (Hooshangi and Alesheikh, 2017). Every
individual robot frequently observes the status of the neighbour
robots within its field of view (Sung et al.,2018). It compares its
task execution competency from its neighbours’competency
and self-allocates profitable tasks (Lerman et al.,2006;
The current issue and full text archive of this journal is available on Emerald
Insight at: https://www.emerald.com/insight/0143-991X.htm
Industrial Robot:the internationaljournal of robotics research and application
47/6 (2020) 929–942
© Emerald Publishing Limited [ISSN 0143-991X]
[DOI 10.1108/IR-04-2020-0073]
Received 9 April 2020
Revised 9 July 2020
18 August 2020
Accepted 19 August 2020
929
Luo et al., 2014). This method is suitable for a large team of
robots in weak communication environments (Wang et al.,
2018). Though the centralized coordination method
necessitates a rigid communication network, it ensures the
consensus about task allocation among the team of robots.
Whereas, distributed coordination is unsusceptible to frail
communication, but the consensus of task allocation is difficult
to achieve (Giordani et al.,2010). The responsibilities of
individual robots in MRS networks are task execution and
coordination (Irfan and Farooq, 2016;Xie et al.,2018). The
coordination methods have their own advantages and
disadvantages. Thus, the proper selection of a coordination
method for an application influences successful task allocation
and task accomplishment in an MRS.
Multi-robot task allocation (MRTA) problems are
categorized into eight types, as shown in Figure 1 (Gerkey and
Matari
c2004). Korsah et al. (2013) group them into four
(Figure 2) based on inter-dependent resources and constraints
and refer them as iTax classification. The task allocation
strategies are classified with respect to the consequential
applications such as search and rescue, surveillance, foraging,
flocking, formation, target tracking, cooperative manipulation
and exploration (Jia and Meng, 2013;Darmanin and Bugeja,
2017;Jin et al., 2019). Two important task allocation strategies
are auction- and optimization-based techniques (Badreldin
et al.,2013;Khamis et al., 2015). The evaluation of these task
allocation strategies based on the solution optimality, allocation
time and problem constraints determine that optimization-
based task allocation is faster and produces optimal solutions
for complex constrained problems.
In recent years, the researchers have focused on developing
dynamic task allocation strategies for complex constraint
problems and developing robust strategies with multiple
uncertainty conditions. This paper aims to survey the
contemporary dynamic task allocation strategies (Nunes et al.,
2017). It recognizes the expedient task allocation strategies for
variant real-world MRS applications. It performs a review of
task allocation strategies by analyzing the problem applications,
constraints, objective functions, task allocation and completion
time and the uncertainty handling methods. In this paper, the
authors have attempted to review, categorize and evaluate the
related papers to provide a systematic view of past work and
provide various research scopes in this problem domain. This
study reviewed published articles from 2010–2020, from high-
ranking journals and reputable international conferences, most
of which were related to multiple-robot task allocation, swarm
robots, scheduling and optimization methods and were
extracted from the “Web of Science and Scopus”databases.
The performance of various strategies is compared considering
the holistic nature of the problem and parameters such as
number of robots, tasks, time for task allocation and
completion, uncertainty conditions and several other
constraints.
This paper is organized as follows: Section 2 defines the
multiple-robot dynamic task allocation problem. Section 3
presents a detailed analysis of four distinct task allocation
strategies. Section 4 provides a detailed discussion of the
analyzed literature, and Section 5 presents the possible research
gaps and scope in this area. Section 6 concludes the major
findings in this paper.
2. Multi-robot task allocation problem definition
This section outlines the MRTA problem. Let J={j
1
,j
2
,j
3
...j
m
}
be the set of tasks to be allocated, and R={r
1
,r
2
,r
3
...r
n
}betheset
of robots in the team. The term Ain equation (1) represents that
the set of tasks Jare assigned to the set of robots R:
A:J!R(1)
If a task jis allocated to a robot r, then task allocation A
j,r=
1
else A
j,r=
0.
Let U
J
[Rbe a matrix of required utility values to execute m
tasks by nrobots.
Let U
R
[Rbe a matrix of available utility values to execute m
tasks by nrobots.
Let T={T
1
,T
2
,T
3
,...T
m
} be the start time of mtasks.
Let W = {W
1
,W
2
,W
3
,...W
m
} be the waiting time of the
tasks to commence.
Let M={M
1
,M
2
,M
3
,...M
m
} be the set of time span of the
tasks.
Let D
j,r
be the distance travelled by the robot rto execute the
task j.
Let K={K
1
,K
2
...K
l
} be the set of completed tasks.
Task assignment to the robots is an optimal decision-making
problem. It is subject to some essential constraints. The various
Figure 1 Fundamental MRTA taxonomy
Figure 2 iTax MRTA taxonomy
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
930
multi-robot problem constraints are listed in Table 1. The basic
dynamic task allocation problem consists of resource and time
constraints as stated below:
A task must be allocated to a robot with sufficient utility,
as depicted in equation (2):
If Uj>Urthen Ajr ¼0;8j2J and r 2R(2)
The task execution time Tof the tasks allocated to a robot
must not overlap with each other. Equation (3) depicts
this constraint. It includes starting time, waiting time and
time span of the tasks:
TjpTjq1Wjq1Mjq
where;jpand jq
e
J and jpjq;
the task jpmust be scheduled after the task jq(3)
Thetaskassignmentmustbeconflict-free. Equation
(4) specifies that a task must be allocated to a single
robot:
Xr2RTj;r¼18j2J(4)
The most common objective functions (equation (5)–(7))of
dynamic task allocation strategy are: minimize the travel
distance (d), the waiting time (W) and to maximize the task
completion rate (K):
min :Xj2JXr2RDj;r(5)
minXj2JWj(6)
max :Xl2JjKljlm(7)
This multi-robot dynamic task allocation problem is a
combinatorial optimization problem. In this paper, the existing
strategies for solving this problem are comprehensively
reviewed and will be discussed in detail in the remaining
sections of the paper.
3. Dynamic task allocation strategies
The objective of multi-robot dynamic task allocation problems
is mapping of the tasks with the robots by satisfying the
constraints such a way that the cost function is minimal (Sarkar
et al., 2018a). The prominent part of real-world multi-robot
applications is that the constraints are irregular and diverse in
nature. Hence, a rational task allocation strategy necessitates
robust handling of the distinct constraints occurs at the time of
task execution. There are various task allocation methods
reported in the literature, and Figure 3 shows the four broad
classifications of multi-robot dynamic task allocation strategies.
This section discusses the task allocation strategies in detail.
In this study, the concepts of numerous MRTA strategies
form the literature are discussed. The discussion is based on the
problem application, the objective function, additional
constraints involved, coordination type, the problem
taxonomy, task reallocation feature, uncertainty handling
method. In addition to these points, the number of tasks and
robots considered in the implementation stage, the resultant
average task allocation time, task completion time and the
implementation method, either simulation or real
experimentation, are considered. This way of analysis helps the
reader to identify the suitable strategy to be considered for the
problem scenario.
3.1 Market-based task allocation
Market-based task allocation is a prominent multiple-robot
task allocation strategy (Schneider et al., 2017). It imitates the
market trading concept (Luo et al.,2015). The process of
market-based task allocation strategy is illustrated in Figure 4.
An auctioneer robot advertises tasks information to other
robots in the team and requests for bids. Every individual robot
Table 1 Dynamic task allocation constraints with examples
Constraints Types
Environmental constraints Moving obstacles, unknown environment, cluttered environment, etc.
Robot constraints Sensor malfunction, communication loss, uncertainty of robot’s travel distance, heterogeneity draining of battery capacity,
computational capacity, resource constraint (Chen and Sun, 2011;Notomista et al., 2019;Schillinger et al., 2019)
Task constraints Time-bounded tasks, multi-agent tasks, hierarchical tasks (Blankenburg et al., 2017), task variants (Cano et al., 2018)
Figure 3 Classification of dynamic task allocation strategies
Figure 4 Process of market-based task allocation
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
931
in the team prepares bid based on its capability to execute the
tasks and then forwards the bids to the auctioneer robot. The
auctioneer robot allocates the tasks to the least quoted robots.
This strategy is implemented in both centralized and
distributed robot’s coordination network (Schneider et al.,
2017).
Market-based task allocation is broadly divided into single-
item auctioning and combinatorial auctioning. The single-item
auctioning (Otte et al.,2020) method conducts task-wise
auctions. Whereas, combinatorial auctioning conducts
auctions for a group of tasks. Researchers develop multiple
variants of these two methods. Table 2 illustrates a detailed
discussion of various market-based task allocation strategies
from the literature. To handle uncertainty, the task allocation
list and cost estimate are periodically updated. With respect to
the updated cost estimate, the tasks are reallocated (Turner
et al., 2017;Turner, 2018). Search and rescue tasks incorporate
several uncertainties such as risk level of the victims, cluttered
path and robot energy level (Talebpour and Martinoli, 2018).
The interval uncertainty theory handles these uncertainties. It
updates the bid estimates within a periodical interval and
reallocates the tasks. This approach improves the task
completion rate and increases the number of life-saved victims
in search and rescue applications (Hooshangi and Alesheikh,
2017). However, task reallocation handles uncertainty and
improves robustness, task switching increases computation
time and it indirectly increases the robot energy consumption
(Talebpour and Martinoli, 2018). The procedural process of
auctioning consumes much time. In common, all the market-
based task allocation strategies result in less robot travel
distance. The market-based task allocation strategies rely on
strongly connected robots’networks. The task completion rate
of market-based strategies during communication loss or weak
communication environments is poor.
The variants of single-item auctioning and combinatorial
auctioning are found in the literature. It supports all kinds of
coordination. Most of the authors used single-objective
functions, and it is observed that task allocation strategies are
implemented in simulation than in real experiments as can be
seen (S shows simulation, R shows real-time experiments).
There exists a large gap to identify the difference between the
simulation results and the real-time execution results.
3.2 Optimization-based task allocation
Real-world multiple robot problems are bounded with multiple
uncertain constraints. Therefore, the mathematical modelling
of dynamic task allocation problems is formidable (Li and
Yang, 2018). However, heuristic modelling of dynamic task
allocation problems furnishes contiguous optimal solutions.
The multiple mobile robots’dynamic task allocation problem is
generalized as multiple travelling salesman problem (Arif and
Haider, 2017) and classified as a combinatorial optimization
problem (dos Reis and Bastos, 2017). It is solved by
evolutionary optimization algorithms: genetic algorithm (Arif
and Haider, 2017;Arif and Haider, 2018;Bänziger et al.,
2018), particle swarm optimization (PSO) (Alshawi and
Shalan, 2017), ant colony optimization (ACO) (Li et al.,
2017b), the variants of PSO and ACO algorithms (Muhuri and
Rauniyar, 2017).
The essential objective functions of dynamic task allocation
problems are minimization of: task completion time, robot
travel distance, battery resource utilization; maximization of:
task distribution rate and task completion rate. Table 3 lists the
features of various optimization strategies. Similar to market-
based strategy, most of the optimization strategies consider
single-objective optimization only. Several methods have been
used in the literature for this. Few researchers considered
integer programming (Li and Li, 2017;Su et al., 2018;Zhou
et al.,2019) and various search algorithms (Zhao et al.,2015;
Kartal et al.,2016;Mitiche et al.,2019). Many researchers used
metaheuristic algorithms (Liu and Kroll, 2012;Alshawi and
Shalan, 2017;Li et al., 2017b,Z.Zhu et al.,2017;Arif and
Haider, 2018;Chen et al., 2018b,Padmanabhan Panchu et al.,
2018;Wang et al., 2018;Zhou et al., 2019) to solve this
optimization problem, and this could be because of the ease of
implementation of such algorithms.
Generation of multiple solutions for a problem instance
enhances the robustness of single-objective optimization
(Huang et al.,2018). Though this technique theoretically
enhances the robustness, there is a gap for systematic switching
between multiple solutions. An optimization strategy consumes
high computational resources (Shenoy and Anupama, 2017).
Multi-objective optimization approach improves the task
completion rate with less time utility. The issue in multi-
objective optimization is framing the proper fitness function. As
some of the objective factors of dynamic task allocation have a
functional trade-off, the weight value of each objective factor
must be precisely given. In future, it is recommended to
perform a study on regularity and adoption of fitness factors for
distinct task allocation problems.
The task distribution rate of multi-objective optimization is
higher than the single-objective optimization strategy. This
approach has better scalability. However, the robustness is low
because of the poor adaptability to multiple performance
objective factors. However, there is an open research gap to
identify differences between the simulation and real
implementation of optimization-based dynamic task allocation.
3.3 Behaviour-based task allocation
Behaviour-based dynamic task allocation is a unique strategy.
This strategy exerts multiple prorated solutions to solve distinct
problem instances taking place in a single application. The
problem solutions are in any form of mathematical model,
heuristic or optimization functions. This strategy is highly
reactive to the problem. Multi-robot exploration problems
consist of two-layered behaviour-based control architecture
(Chetty et al., 2010;Chetty et al.,2011). Tasks identification
and inter robot communication are classified under higher-level
behaviours, whereas obstacle avoidance, navigation and task
switching are categorized into lower-level behaviours. In
addition to the basic low-level behaviours, problem-specific
behaviours are developed to incorporate robustness
(Schillinger et al., 2018).
Table 4 summarizes the analysis of various behaviour-based
task allocation techniques. Unlike other task allocation
strategies, this approach is adaptable and reactive to the
problem’s specific constraints. Thus, this strategy leads to high
robustness and scalability features. This approach is adaptable
for centralized and distributed coordination. Distributed
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
932
Table 2 Analysis of market-based task allocation
Source Application Method Objective function
Additional
constraint
Problem
type Coordination Reallocation Uncertainty
Maximum
no.: tasks
Maximum
no.: robots
Average task
allocation time (s)
Average task
completion time
Real time/
simulation
Chen and Sun (2011)Generic Leader–follower
coalition algorithm
Maximize utility Heterogeneous team
with resource
constraint
ST-SR-IA Distributed Y N 5 3 ––S
Luo et al. (2012)Generic Repeated greedy auction
algorithm
Maximize total payoff Task group ST-SR-TA Distributed N N 60 20 ––S
Liu and Shell (2012)Generic Optimal auctioning with
strategic pricing
Minimize distance travelled –ST-SR-IA Distributed Y N 500 500 0.5 –S
Tolmidis and Petrou
(2013)
Generic Hybrid genetic algorithm
(GA) distributed auction
Maximize robot battery
energy and relevance
degree, minimize time
Remaining energy
has to be above a
predefined level
ST-SR-IA Distributed Y N 10 6 5.62 –S
Liekna et al. (2012)Multiple vacuum
cleaning robots
Contract Net protocol Minimize the effort
required for cleaning an
area
–SR-MT-IA Centralized Y N 2 2 ––S
Liu and Shell (2013)Generic Linear integer
programming with
partitioning of tasks
Maximize profit–ST-SR-IA Distributed N N 100 100 ––S
Luo et al. (2015)Generic Iterative auctioning Maximize remaining
battery power
Task deadline ST-SR-TA Distributed N N 100 20 ––S
Wei et al. (2016)Search and retrieval SSI. Extend SSI Minimize the completion
time and fuel consumption
Temporal constraint ST-SR-IA Centralized/
distributed
N N 30 10 –150.57 (s) S
Farinelli et al. (2017)Multi-robot patrolling SSI auctions Maximize the number of
visits
–ST-SR-IA Distributed N Y 8 3 ––R
Hooshangi and
Alesheikh (2017)
Search and rescue Contract Net protocol Maximize the number of
rescued victims
Heterogeneous team ST-SR-IA Distributed Y Y 2000 200 –738 (min) S
Otte et al. (2020)Lossy communication
environment
Comparison of six
auction algorithms
Minimize the maximum
path length
Communication
limited environments
ST-SR-IA Distributed N N 1000 300 ––S
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
933
Table 3 Analysis of optimization-based task allocation
Source Application Method Objective function
Additional
constraint
Problem
type Coordination Reallocation Uncertainty
Maximum
no.: tasks
Maximum
no.: robots
Average task
allocation time (s)
Average task
completion time (s)
Real time/
simulation
Wawerla and Vaughan
(2010)
Puck transportation Centralized planner and
heuristic rules
Minimize robot energy
consumption
–ST-SR-IA Centralized/
distributed
YN218 ––S
Jevtic et al. (2011)Generic Distributed bee colony
optimization
Maximize task distribution –ST-SR-IA Distributed N N 4 100 ––S
Liu and Kroll (2012)Industrial plant
inspection
Aand GAs Maximize task completion –ST-SR-IA Centralized N N 90 3 170.48 –S
Wang et al. (2012)Generic ACO Minimize travel distance Resource constraints ST-SR-IA Distributed N N –– – – S
Giordani et al. (2013)Industry production Iterative auction-based
negotiation
Minimize production cost –ST-SR-IA Distributed Y N 50 250 ––S
Zhao et al. (2015)Search and rescue Heuristicbased Minimize sum of path cost Limited resources MT-MR-IA Distributed Y N 32 16 –238.49 S
Kartal et al. (2016)Generic Monte Carlo tree search Minimize travel distance ST-SR-TA Centralized N N –S
Arif and Haider (2017)Generic GA Minimize travel distance –ST-SR-IA Centralized N N 30 3 –41.54 S
Alshawi and Shalan
(2017)
Foraging PSO Minimize time –ST-SR-IA Distributed N N 10 7 14.6 –S
Z. Zhu et al. (2017)Generic Improved PSO Maximize benefit with
minimum travelling
distance and paid cost
Payload constraint ST-SR-IA Distributed N N 40 6 5208 –S
Li and Li (2017)Warehouse
automation
Integer programming
and GA
Minimize the sum of the
fixed cost of robot and the
cost of robot operation
–ST-SR-IA Centralized Y N 10 10 –S
Li et al. (2017b)Generic ACO Minimize travel distance –ST-SR-IA Distributed N N 10 –– – S
Tsang et al. (2018)Warehouse
automation
GA Minimize travel distance –MT-MR-IA Centralized N N 100 100 0.340 –S
Turner et al. (2017)Search and Rescue PI-MaxAss Maximize the number of
task allocations
Time and fuel limit ST-SR-TA Distributed Y N –– – – S
Chen et al. (2018a)Generic PSO multi-objective Maximize time utility and
energy utility
–ST-SR-IA Distributed N N 16 16 ––S
Mitiche et al. (2019)Generic Iterated local search Maximize the number of
tasks
Spatio-temporal and
capacity
ST-SR-TA Distributed N N –– – S
Zhou et al. (2019)Generic Integer programming
and approximation tree-
based GA
Minimize task completion
time
–ST-SR-IA Centralized N N –– – – S
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
934
Table 4 Analysis of behaviour-based task allocation
Source Application Method Behaviour Objective function Additional constraint
Problem
type Coordination Reallocation Uncertainty
Maximum no.:
tasks
Maximum
no.: robots
Average task
allocation time
(s)
Average task
completion
time (s)
Real time/
simulation
Chen and Sun (2012)Generic Sequential coalition
method
Coalition Maximize coalition utility Resource constraint MR-ST-IA Distributed N –610––R
Lee et al. (2014)Foraging Iterative search on ad
hoc network
Resource aware cost
generation
Minimize resource
consumption
Limited resources ST-SR-IA Distributed Y 100 16 –64.9 S
Kanakia et al. (2016)Generic Game theory Bayesian
Nash equilibrium
Continuous response
threshold
Maximize task completion No communication ST-SR-IA Distributed N Communication ––––S
Riccio et al. (2016)Soccer game/foraging Distributed world
modelling and task
allocation
Context knowledge
based
Minimize time –ST-SR-IA Distributed N –13––R
Abukhalil et al. (2016)Search and Rescue Robot utility-based task
assignment
Robot utility-based
allocation
Maximize utility Heterogeneous team ST-MR-IA Centralized/
distributed
Y–51–110.3 R
Lee and Kim (2017)Foraging Task selection probability
model
Response threshold
behaviour
Maximize task distribution No communication ST-SR-IA Distributed Y –50 20 ––S
Wu et al. (2017)Generic Gini coefficient and
auction-based allocation
Gini coefficient-based
allocation
Minimize resource
consumption
Limited energy resources ST-SR-IA Centralized N Resource
constraint
50 5 ––S
Lee (2018) Goods delivery mission Probabilistic bid
auctioning
Resource-based task
allocation
Minimize the maximum cost
and time
Fuel refill station ST-SR-IA Distributed Y Resource-level
uncertainty
72 11 ––S
Talebpour and Martinoli
(2018)
Pedestrian walking Adaptive risk-based re-
planning strategy
Risk-based allocation Minimize travel distance Social constraints ST-SR-IA Distributed Y Human walking 5 4 ––R
Dai et al. (2019)Soccer game Incomplete information
game modelling
Ball velocity-based
allocation
Minimize the payoff No communication ST-SR-IA Distributed N ––3––R
Jin et al. (2019)Target tracking Competition-based task
allocation
Besieging behaviour-
based allocation
Maximize task completion Limited communication ST-MR-IA Distributed N –––––S
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
935
coordination requires local communication rather than global
communication between robots for task allocation. Therefore,
this strategy is suitable for weak communication applications.
Task switching or swapping behaviour is incorporated with the
task allocation model. Thus, this strategy could handle a
robot’s failure that occurred during the task execution phase
(Zhao et al.,2015). The resource-based task allocation
behaviour provides an improved task distribution rate (Lee and
Kim, 2019). Therefore, behaviour-based task allocation is
recommended for task allocation in uncertain and dynamic
real-world multiple mobile robot (MMR) applications.
3.4 Clustered task allocation
Clustered task allocation strategies group the similar or nearby
tasks into clusters, and then the clusters are allocated to robots
rather than single task allocation. This strategy decreases the
average travel distance of the robot team (Chen et al., 2018b).
For search and rescue applications, nearby tasks are clustered.
Similarly, for warehouse operations, nearby tasks clustering is
implemented. According to Sarkar et al. (2018b), as the robots
execute the nearby tasks, the overall travel distance and the
resource utilization are less. Similar types of tasks are clustered
for heterogeneous robots’teams, and the consent task type is
allocated to the corresponding type of robots. For the clustering
of tasks, methods such as Euclidean distance clustering, K
means clustering, fuzzy clustering (Ghassemi and Chowdhury,
2018) are reported to be used in the literature.
The critical challenge in clustering-based task allocation is
the identification of the optimal number of tasks per cluster
(Nam and Shell, 2016). The determination of optimal number
of tasks for a cluster is empirically studied in Mitiche et al.
(2019). The performance analysis of clustering-based task
allocation approaches is given in Table 5. A task clustering
strategy minimizes the travel distance; thus, it is recommended
for foraging applications. Clustered task allocation reduces the
number of individual tasks to be allocated. Therefore,
computational complexity is drastically reduced in this
approach. This strategy is adaptable with centralized
coordination. Thus, it is not relevant for weak communication
applications. As a group of tasks is assigned to a robot, the
occurrence of a robot’s failure will decrease the task completion
rate drastically. This downside is solved by task switching/
swapping with the next available robots in case of robot failure
conditions.
4. Discussion
This paper presents a detailed review of the state-of-the-art
MMR dynamic task allocation strategies. Even though every
task allocation technique in the literature has been analyzed and
validated by simulations, it lacks validation through real-time
experimentation. Identification of the gaps between simulation
and experimental results is still open research in the field of
dynamic task allocation (Jang et al.,2018). In market-based
task allocation, robots have the provision to opt for profitable
tasks. This method results in the optimal selection of
subsequent tasks. Conventionally, sequential single-item (SSI)
and parallel auctioning require a prior list of tasks. Repeated
auctioning is preferred for dynamic task allocation problems
because it updates tasks in a dynamic manner, in turn
eliminating the requirement for the prior tasks list. Under
uncertain conditions, it is plausible to lose communication
within the robot’s team. From survey, it is found that a robust
and reliable communication network is mandatory for the
market-based task allocation leading to an open challenge that
is not yet been solved. Relying on a single-point auctioneer for
task allocation is the major downside of this strategy.
Development of a systematic strategy to handle this single-
point failure is a further potential research direction in market-
based task allocation. Various optimization algorithms based
on dynamic task allocation strategies are reported in the
literature. Task allocation quality relies on the fitness function
used in the optimization algorithms. Overall, the optimization-
based task allocations provide less robustness. Thus,
determination of self-adaptive fitness function to successfully
handle uncertainties is yet to be considered by the researchers.
The analysis of the literature states that multi-objective
optimization techniques outperform single-objective dynamic
task allocation. However, there exists a huge research gap in
deriving multi-objective cost function with the trading-off
objective factors. Several heuristic algorithms are available;
thus, further research is recommended for identification and
validation of optimal heuristic technique for specific multi-
robot problems. Implementation and performance comparison
of various optimization algorithms is another research direction
that could be performed.
The uniqueness of behaviour-based task allocation strategy
makes it flexible to incorporate reactive behaviours of robots for
various problem constraints and uncertainties. This feature
increases the robustness and scalability of task allocation.
Arbitration among multiple behaviours is the challenging
aspect of this strategy. This strategy consumes more
computation resources, which is another drawback. However,
this strategy is recommended for applications with various
uncertainties.
Task clustering-based allocation strategy consumes minimal
travel distance for the robots. This strategy is suitable for
autonomous multi-robot surveillance applications.
Identification of an optimal number of tasks in a cluster is the
research problem open for further study. Simultaneous task
allocation and path planning strategy increase the task
completion rate (Sung et al.,2018). Development of effective
switching strategies between clusters to handle uncertainty like
robot failures is the further research direction in this strategy.
Similarly, integrating the different task allocation strategies like
behaviour-based allocation with task clustering-based task
allocation strategy is also a significant research direction for the
future. Table 6 illustrates the performance factors for MRTA
problems. The ways of adapting to these factors by the four task
allocation strategies are presented in detail.
5. Gaps and future research scope
From the literature review analysis, it is identified that a
multiple-robot system with different depot points for each
robot decreases the total travel distance and inter-robot
collisions (Lu et al., 2018). Task allocation for complex
constraint problems, including time window tasks (Liu et al.,
2017), hierarchical tasks (Blankenburg et al., 2017), robot
dependent tasks, task unknown problems, is still open for
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
936
Table 5 Analysis of clustered task allocation
Source Application Method Objective function
Additional
constraint
Problem
type Coordination Reallocation Uncertainty
Maximum
no.: tasks
Maximum
no.: robots
Average task
allocation
time (s)
Average task
completion
time (s)
Real time/
simulation
Zhang et al. (2012)Generic Stochastic clustering
auction, constrained
Prim’s algorithm
Minimize time –MT-SR-IA Centralized/
distributed
NN–6––S
Faigl et al. (2012)Exploration Multiple travelling
salesmen-based
assignment; K-means
clustering
Minimize travel distance –MT-SR-IA Distributed N N –10 –– S
Biswas et al. (2017)Generic PSO with k-means
clustering
Minimize travel distance –MT-SR-TA Distributed N N 10 3 –– S
Chen et al. (2018b)Search and rescue Cluster first consensus-
based strategy
Maximize the no.: of
rescued survivors,
minimize the waiting
time of survivors, the
total travelling distance
of robots
–MT-SR-IA Distributed N N 2 14 ––S
Lu et al. (2018)Foraging Central place foraging
algorithm, k-means
clustering
Minimize travel distance –MT-MR-IA Distributed N N 384 24 –– S
Sarkar et al. (2018b)Warehouse Nearest neighbour-based
clustering and routing
Minimize travel distance Robot capacity
constraint
MT-SR-IA Distributed N N –– – – S
Ghassemi and
Chowdhury (2018)
Generic Fuzzy clustering, bipartite
graph matching
Minimize travel distance –MT-SR-IA Distributed N N 100 50 4.63 S
Whitbrook et al.
(2019)
Generic Robust performance
impact algorithm
Minimize mean
individual task cost
–MT-SR-IA Distributed N Y 32 16 S
Dutta et al. (2019)Generic Linear programming-
based graph partitioning
Maximize coalition
structure, after
minimizing the cost of
forming it
–MT-MR-IA Centralized N N 10 100 230 –S
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
937
research (Khamis et al.,2015). Development of soft agents for
robust task allocation is also a further research direction that
can be explored (Ismail and Sariff, 2018). Development of the
game theoretical approach for distributed task allocation for
communication loss problems is recommended for further
research analysis (Dai et al.,2019). A very few researchers
identified dynamic task allocation strategies by applying some
modern approaches such as learning automata (Khani et al.,
2019), deep learning, machine vision (Li and Yang, 2018), self-
organizing map neural networks (Zhu et al., 2017).
Consider the following examples of MRTA problems,
dynamic task allocation for search and rescue of victims by
autonomous heterogeneous MMR in disaster field problem is a
highly challenging problem for the researchers (Abukhalil et al.,
2016). The list of search and rescue tasks to be performed is
unknown in prior. Thus, the task allocation strategy for this
problem must be scalable for tasks as well as robots during
runtime. Rescue tasks must be given higher priority than the
search tasks (Sung et al.,2018). Priority of the behaviour-based
task allocation strategy ensures the execution of rescue tasks
prior to the search tasks. Integration of task clustering-based
allocation reduces the robot travel distance. This problem
contains an uncertain environment and obstacles. Thus, it is
recommended to develop a task allocation strategy that is
reactive for the uncertain behaviours of obstacles (ElGibreen
and Youcef-Toumi, 2019). The distributive task allocation
strategy is applicable to this problem because strong
communication may not exist in all the disaster fields.
Incorporating task switching and reallocation mechanisms in
case of robot failures and other uncertainties improves the
robustness of task allocation (Chen et al., 2016;Woosley and
Dasgupta, 2018). Another challenge in this problem is the
proper dynamic task allocation and coordination between
heterogeneous mobile robots. Modelling the dynamic task
allocation problem in a three-dimensional environment is
efficacious to manipulate real-time application environment
(Yi et al.,2016). In multi-robots gaming applications, the
unpredictable changes in the game environment are the
challenging aspect for task allocation. Similar to these
scenarios, every real-world multiple robot application places
itself with distinct challenges and uncertainties for task
allocation. The performance comparison of different task
allocation strategies in real-time is a significant research
direction (Ismail and Sariff, 2018). In Table 7, various future
research scopes in MRTA problems are summarized for the
reader’s clarity.
6. Conclusion
This paper analyzed in detail several dynamic task allocation
strategies developed and reported for multiple-robot
systems. Even though many of the proposed strategies are
Table 6 Factors for MRTA problems
Performance factors Market-based allocation
Optimization-based
allocation Behaviour-based allocation
Task clustering-based
allocation
No communication/lossy
communication
Multiple times broadcasting of
winner robot details after
bidding
Local communication among
neighbour robots
Local communication among
neighbour robots
Local communication among
neighbour robots
Objective function Single/multiple objective Single/multiple objective Single/multiple objective Single/multiple objective
Coordination type Centralized/distributed Centralized/distributed Centralized/distributed Centralized/distributed
Method for task
reallocation
Iterative auctioning methods Iterative searching and
allocation
Heuristics rules searching/
Bayesian Nash equilibrium
Difficult reallocate within
different task clusters
Uncertainty handling
techniques
Iterative auctioning methods Difficult to handle
uncertainties
Game theory/probabilistic
predictive modelling
Difficult to handle uncertainty
Complex problem
constraints
Difficult to conduct auctions Complex and difficult to solve
due to multiple decision
variables
Can be handled in a collective
manner
Can be handled in a collective
manner
Computational cost Lower than optimization
strategy
Higher than market-based
strategy
Higher than optimization-
based strategy
Lower than other methods
Table 7 Future research avenues in MRTA problems
Development of task allocation strategies for tightly coupled tasks
Development of efficient failure handling mechanisms for all types of task
allocation
Development of hybrid task allocation strategies
Development of energy-aware task allocation strategies for recharging
robots to perform long-running tasks
Evaluation of task priorities in task allocation
Unifying the performance evaluation metrics of task allocation strategies
Development of task allocation for strict time-constraint problems
Development of effective task rescheduling/reallocation mechanism
Development of task allocation within a heterogeneous team
Development of task allocation for unknown environment
exploration problems
Development of adaptive heuristic parameters for task allocation
Identification of ideal cluster size and procedure for task switching
among task clusters
Development of task allocation in no communication problem
environment
Development of uncertainty handling task allocation techniques
Real-time experimentation of task allocation problems
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
938
validated using simulations study, very few strategies are
tested in real time. There exists a large gap between the
simulated and real-time multi-robot applications. Further
research needs to be conducted exclusively to overcome the
experimental challenges of real-world applications. The
multi-robot systems are prone to communication
uncertainty; thus, the current dynamic task allocation
techniques intend to be improved for achieving robustness
in weak or no communication scenarios. Robust task
allocation with efficienttaskswitchingandswapping
techniques to manipulate the uncertainties is a needful
research direction in multi-robot systems. Multiple
behaviour-based dynamic task allocation techniques
improve the scalability and robustness. Another major
challenge is the development of dynamic task allocation
strategies for exploration problems because an inadequate
number of studies have been reported on this problem. The
major challenge exists in the obscure knowledge of tasks to
be allocated in the unperceived application environment.
The exploration problem entails task allocation in parallel
with the path planning of an unknown application
environment. Hence, a successful exploration task
allocation strategy is required to be scalable for tasks and to
be robust for environment uncertainties. Researchers can
also develop integrated and robust behaviour-based
dynamic task allocation strategies for search and rescue
applications as future work.
References
Abukhalil, T., Patil, S. Patel, M. and Sobh, T. (2016),
“Coordinating a heterogeneous robot swarm using robot
utility-based task assignment (RUTA)”,in: 2016 IEEE 14th
International Workshop on Advanced Motion Control (AMC),
IEEE.
Alshawi, M.A. and Shalan, M.B. (2017), “Minimal time
dynamic task allocation for a swarm of robots”,International
Journal of Mechanical Engineering and Robotics Research,
Vol. 6 No. 6.
Arif, M.U. and Haider, S. (2017), “An evolutionary traveling
salesman approach for Multi-Robot task allocation”, in:
Icaart (2).
Arif, M.U. and Haider, S. (2018), “Aflexible evolutionary
algorithm for task allocation in multi-robot team”,in:
International Conference on Computational Collective
Intelligence,Springer.
Badreldin, M., Hussein, A. and Khamis, A. (2013), “A
comparative study between optimization and Market-Based
approaches to Multi-Robot task allocation”,Advances in
Artificial Intelligence, Vol. 2013(16877470).
Bänziger, T., Kunz, A. and Wegener, K. (2018), “Optimizing
human–robot task allocation using a simulation tool based
on standardized work descriptions”,Journal of Intelligent
Manufacturing, Vol. 31 No. 7.
Biswas, S., Anavatti, S.G. and Garratt, M.A. (2017), “Nearest
neighbour based task allocation with multi-agent path
planning in dynamic environments”,in: 2017 International
Conference on Advanced Mechatronics, Intelligent Manufacture,
and Industrial Automation (ICAMIMIA),IEEE.
Blankenburg, J., Banisetty, S.B., Alinodehi, S.P.H., Fraser, L.,
Feil-Seifer, D., Nicolescu, M. and Nicolescu, M. (2017), “A
distributed control architecture for collaborative multi-robot
task allocation”,in: 2017 IEEE-RAS 17th International
Conference on Humanoid Robotics (Humanoids),IEEE.
Cano, J., White, D.R., Bordallo, A., McCreesh, C., Michala,
A.L., Singer, J. and Nagarajan, V. (2018), “Solving the task
variant allocation problem in distributed robotics”,
Autonomous Robots, Vol. 42 No. 7, pp. 1477-1495.
Chen, J. and Sun, D. (2011), “Resource constrained
multirobot task allocation based on leader–follower coalition
methodology”,The International Journal of Robotics Research,
Vol. 30 No. 12, pp. 1423-1434.
Chen, J. and Sun, D. (2012), “Coalition-based approach to
task allocation of multiple robots with resource constraints”,
IEEE Transactions on Automation Science and Engineering,
Vol. 9 No. 3, pp. 516-528.
Chen, J., Wang, J., Xiao, Q. and Chen, C. (2018a), “A Multi-
Robot task allocation method based on Multi-Objective
optimization”, in: 2018 15th International Conference on
Control, Automation, Robotics and Vision (ICARCV),IEEE.
Chen, X., Zhang, P., Li,F. and Du, G. (2018b), “A cluster first
strategy for distributed multi-robot task allocation problem
with time constraints”, in: 2018 WRC Symposium on
Advanced Robotics and Automation (WRC SARA),IEEE.
Chen, Y., Mao, X., Hou, F., Wang, Q. and Yang, S. (2016),
“Combining re-allocating and re-scheduling for dynamic
multi-robot task allocation”,in: 2016 IEEE International
Conference on Systems, Man, and Cybernetics (SMC),IEEE.
Chetty, R.K., Singaperumal, M. and Nagarajan, T. (2010),
“Behaviour based planning and control of leader follower
formations in wheeled mobile robots”,International Journal
of Advanced Mechatronic Systems, Vol. 2 No. 4, pp. 281-296.
Chetty, R.K., Singaperumal, M., Nagarajan, T. and Tetsunari,
I. (2011), “Coordination control of wheeled mobile robots?
A hybrid approach”,International Journal of Computer
Applications in Technology, Vol. 41 Nos 3/4, pp. 195-204.
Dai, W., Lu, H., Xiao, J. and Zheng, Z. (2019), “Task
allocation without communication based on incomplete
information game theory for multi-robot systems”,Journal of
Intelligent & Robotic Systems, Vol. 94 Nos 3/4, pp. 841-856.
Darmanin, R.N. and Bugeja, M.K. (2017), “A review on
multi-robot systems categorised by application domain”,in:
2017 25th Mediterranean Conference on Control and
Automation (MED),IEEE.
D’Emidio, M. and Khan, I. (2017), “Multi-robot task
allocation problem: current trends and new ideas”,in: Joint
Proceedings of the 18th Italian Conference on Theoretical
Computer Science and the 32nd Italian Conference on
Computational Logic (ICTCS/CILC).
dos Reis, W.P.N. and Bastos, G.S. (2017), “An arrovian view
on the multi-robot task allocation problem”,in: 2017 18th
International Conference on Advanced Robotics (ICAR),IEEE.
Dutta, A., Ufimtsev, V. and Asaithambi, A. (2019),
“Correlation clustering based coalition formation for multi-
robot task allocation”,in: Proceedings of the 34th ACM/
SIGAPP Symposium on Applied Computing.
ElGibreen, H. and Youcef-Toumi, K. (2019), “Dynamic task
allocation in an uncertain environment with heterogeneous
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
939
multi-agents”,Autonomous Robots, Vol. 43 No. 7,
pp. 1639-1664.
Faigl, J., Kulich, M. and P
reucˇil, L. (2012), “Goal assignment
using distance cost in multi-robot exploration”,in: 2012
IEEE/RSJ International Conference on Intelligent Robots and
Systems,IEEE.
Fang, B., Zhang, Q., Wang, H. and Yuan, X. (2018),
“Personality driven task allocation for emotional robot
team”,International Journal of Machine Learning and
Cybernetics, Vol. 9 No. 12,pp. 1955-1962.
Farinelli, A., Iocchi, L. and Nardi, D. (2017), “Distributed on-
line dynamic task assignment for multi-robot patrolling”,
Autonomous Robots, Vol. 41 No. 6, pp. 1321-1345.
Gerkey, B.P. and Matari
c, M.J. (2004), “A formal analysis and
taxonomy of task allocation in multi-robot systems”,The
International Journal of Robotics Research, Vol. 23 No. 9,
pp. 939-954.
Ghassemi, P. and Chowdhury, S. (2018), “Decentralized task
allocation in multi-robot systems via bipartite graph
matching augmented with fuzzy clustering”,in: International
Design Engineering Technical Conferences and Computers and
Information in Engineering Conference,American Society of
Mechanical Engineers.
Giordani, S., Lujak, M. and Martinelli, F. (2010), “A
distributed algorithm for the multi-robot task allocation
problem”,in: International conference on industrial, engineering
and other applications of applied intelligent systems,Springer.
Giordani, S., Lujak, M. and Martinelli, F. (2013), “A
distributed multi-agent production planning and scheduling
framework for mobile robots”,Computers & Industrial
Engineering, Vol. 64 No.1, pp. 19-30.
Hooshangi, N. and Alesheikh, A.A. (2017), “Agent-based task
allocation under uncertainties in disaster environments: an
approach to interval uncertainty”,International Journal of
Disaster Risk Reduction, Vol. 24, pp. 160-171.
Huang, L., Ding, Y., Zhou, M., Jin, Y. and Hao, K. (2018),
“Multiple-solution optimization strategy for multirobot task
allocation”,IEEE Transactions on Systems, Man, and
Cybernetics: Systems.
Irfan, M. and Farooq, A. (2016), “Auction-based task
allocation scheme for dynamic coalition formations in
limited robotic swarms with heterogeneous capabilities”,in:
2016 International Conference on Intelligent Systems Engineering
(ICISE),IEEE.
Ismail, Z.H. and Sariff, N. (2018), A Survey and Analysis of
Cooperative Multi-Agent Robot Systems: challenges and
Directions, Applications of Mobile Robots, IntechOpen.
Jang, I., Shin, H.-S. and Tsourdos, A. (2018), “Anonymous
hedonic game for task allocation in a large-scale multiple
agent system”,IEEE Transactions on Robotics, Vol. 34 No. 6,
pp. 1534-1548.
Jevtic, A., Gutiérrez, A., Andina, D. and Jamshidi, M. (2011),
“Distributed bees algorithm for task allocation in swarm of
robots”,IEEE Systems Journal, Vol. 6 No. 2, pp. 296-304.
Jha, S.S. and Nair, S.B. (2017), “TANSA: task allocation using
nomadic soft agents for multirobot systems”,IEEE
Transactions on Emerging Topics in Computational Intelligence,
Vol. 2 No. 4, pp. 308-318.
Jia, X. and Meng, M.Q.-H. (2013), “A survey and analysis of
task allocation algorithms in multi-robot systems”,in: 2013
IEEE International Conference on Robotics and biomimetics
(ROBIO),IEEE.
Jin, L., Li, S., La, H.M., Zhang, X. and Hu, B. (2019),
“Dynamic task allocation in multi-robot coordination for
moving target tracking: a distributed approach”,Automatica,
Vol. 100, pp. 75-81.
Johnson, L.B., Choi, H.-L. and How, J.P. (2016), “The role of
information assumptions in decentralized task allocation: a
tutorial”,IEEE Control Systems Magazine, Vol. 36 No. 4,
pp. 45-58.
Kanakia, A., Touri, B. and Correll, N. (2016), “Modeling
multi-robot task allocation with limited information as global
game”,Swarm Intelligence, Vol. 10 No. 2, pp. 147-160.
Kartal, B., Nunes, J. Godoy, E. and Gini, M. (2016), “Monte
Carlo tree search with branch and bound for multi-robot task
allocation”, in: The IJCAI-16 Workshop on Autonomous
Mobile Service Robots.
Khamis, A., Hussein, A. and Elmogy, A. (2015), Multi-Robot
Task Allocation: A Review of the State-of-the-Art. Cooperative
Robots and Sensor Networks, 2015, Springer, pp.31-51.
Khani, M., Ahmadi, A. and Hajary, H. (2019), “Distributed
task allocation in multi-agent environments using cellular
learning automata”,Soft Computing, Vol. 23 No. 4,
pp. 1199-1218.
Korsah, G.A., Stentz, A. and Dias, M.B. (2013), “A
comprehensive taxonomy for multi-robot task allocation”,
The International Journal of Robotics Research, Vol. 32 No. 12,
pp. 1495-1512.
Lee, D.-H. (2018), “Resource-based task allocation for multi-
robot systems”,Robotics and Autonomous Systems, Vol. 103,
pp. 151-161.
Lee, W. and Kim, D. (2017), “History-based response
threshold model for division of labor in multi-agent
systems”,Sensors, Vol. 17 No. 6, p. 1232.
Lee, W. and Kim, D. (2019), “Adaptive approach to regulate
task distribution in swarm robotic systems”,Swarm and
Evolutionary Computation, Vol. 44, pp. 1108-1118.
Lee, D.-H., Zaheer, S.A. and Kim, J.-H. (2014), “Ad hoc
network-based task allocation with resource-aware cost
generation for multirobot systems”,IEEE Transactions on
Industrial Electronics, Vol. 61 No. 12, pp. 6871-6881.
Lerman, K., Jones, C., Galstyan, A. and Matari
c, M.J. (2006),
“Analysis of dynamic task allocation in multi-robot systems”,
The International Journal of Robotics Research,Vol.25No.3,
pp. 225-241.
Li, Z. and Li, X. (2017), “Research on model and algorithm of
task allocation and path planning for multi-robot”,Open
Journal of Applied Sciences, Vol. 7 No. 10, pp. 511.
Li, J. and Yang, F. (2018), “Research on multi-robot
scheduling algorithms based on machine vision”,EURASIP
Journal on Image and Video Processing, Vol. 2018 No. 1,
pp. 1-11.
Li, D., Fan, Q. and Dai, X. (2017a), “Research status of multi-
robot systems task allocation and uncertainty treatment”, in:
J. Phys. Conf. Ser.
Li, X., Liu, Z. and Tan, F. (2017b), “Multi-robot task
allocation based on cloud ant colony algorithm”,in:
International Conference on Neural Information Processing,
Springer.
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
940
Liekna, A., Lavendelis, E. and Grabovskis, A. (2012),
“Experimental analysis of contract net protocol in multi-
robot task allocation”,Applied Computer Systems, Vol. 13
No. 1, pp. 6-14.
Liu, C. and Kroll, A. (2012), “A centralized multi-robot task
allocation for industrial plant inspection by using aand
genetic algorithms”,in: International Conference on Artificial
Intelligence and Soft Computing,Springer.
Liu, L. and Shell, D.A. (2012), “Large-scale multi-robot task
allocation via dynamic partitioning and distribution”,
Autonomous Robots, Vol. 33 No. 3, pp. 291-307.
Liu, L. and Shell, D.A. (2013), Optimal Market-Based Multi-
Robot Task Allocation via Strategic Pricing. in: Robotics,
Science and Systems.
Liu, S., Kurniawan, E., Tan, P.H., Zhang, P., Sun, S. and Ye,
S. (2017), “Dynamic scheduling for heterogeneous
resources with time windows and precedence relation”,in:
TENCON 2017-2017 IEEE Region 10 Conference,IEEE.
Lu, Q., Hecker, J.P. and Moses, M.E. (2018), “Multiple-place
swarm foraging with dynamic depots”,Autonomous Robots,
Vol. 42 No. 4, pp. 909-926.
Luo, L., Chakraborty, N. and Sycara, K. (2012), “Competitive
analysis of repeated greedy auction algorithm for online
multi-robot task assignment”,in: 2012 IEEE International
Conference on Robotics and Automation,IEEE.
Luo, L., Chakraborty, N. and Sycara, K. (2014), “Provably-
good distributed algorithm for constrained multi-robot task
assignment for grouped tasks”,IEEE Transactions on
Robotics, Vol. 31 No. 1, pp. 19-30.
Luo, L., Chakraborty, N. and Sycara, K. (2015), “Distributed
algorithms for multirobot task assignment with task deadline
constraints”,IEEE Transactions on Automation Science and
Engineering, Vol. 12 No.3, pp. 876-888.
Mitiche, H., Boughaci, D. and Gini, M. (2019), “Iterated local
search for time-extended multi-robot task allocation with
spatio-temporal and capacity constraints”,Journal of
Intelligent Systems, Vol. 28 No. 2, pp. 347-360.
Muhuri, P.K. and Rauniyar, A. (2017), “Immigrants based
adaptive genetic algorithms for task allocation in multi-robot
systems”,International Journal of Computational Intelligence
and Applications, Vol. 16 No. 4, p. 1750025.
Nam, C. and Shell, D. (2016), An Empirical Study of Task
Bundling for Sequential Stochastic Tasks in Multi-Robot Task
Allocation, Technical Report TAMU-CSE-16-7-1, CSE
Department, TX A&M University.
Notomista, G., Mayya, S., Hutchinson, S. and Egerstedt, M.
(2019), “An optimal task allocation strategy for
heterogeneous multi-robot systems”, in: 2019 18th European
Control Conference (ECC),IEEE.
Nunes, E., Manner, M., Mitiche, H. and Gini, M. (2017), “A
taxonomy for task allocation problems with temporal and
ordering constraints”,Robotics and Autonomous Systems,
Vol. 90, pp. 55-70.
Otte, M., Kuhlman, M.J. and Sofge, D. (2020), “Auctions for
multi-robot task allocation in communication limited
environments”,Autonomous Robots, Vol. 44 Nos 3/4,
pp. 547-584.
Padmanabhan Panchu, K., Rajmohan, M., Sundar, R. and
Baskaran, R. (2018), “Multi-objective optimisation of multi-
robot task allocation with precedence constraints”,Defence
Science Journal,Vol.68No.2.
Palmer, A.W., Hill, A.J. and Scheding, S.J. (2018), “Modelling
resource contention in multi-robot task allocation problems
with uncertain timing”,in: 2018 IEEE International
Conference on Robotics and Automation (ICRA),IEEE.
Riccio, F., Borzi, E., Gemignani, G. and Nardi, D. (2016),
“Multi-robot search for a moving target: integrating world
modeling, task assignment and context”,in: 2016 IEEE/RSJ
International Conference on Intelligent Robots and Systems
(IROS),IEEE.
Rishwaraj, G. and Ponnambalam, S. (2017), “Integrated trust
based control system for multirobot systems: development
and experimentation in real environment”,Expert Systems
with Applications, Vol. 86, pp. 177-189.
Sarkar, C., Dey, S. and Agarwal, M. (2018a), “Semantic
knowledge driven utility calculation towards efficient multi-
robot task allocation”,in: 2018 IEEE 14th International
Conference on Automation Science and Engineering (CASE),
IEEE.
Sarkar, C., Paul, H.S. and Pal, A. (2018b), “A scalable multi-
robot task allocation algorithm”,in: 2018 IEEE International
Conference on Robotics and Automation (ICRA), IEEE.
Schillinger, P., Bürger, M. and Dimarogonas, D.V. (2018),
“Simultaneous task allocation and planning for temporal
logic goals in heterogeneous multi-robot systems”,The
International Journal of Robotics Research, Vol. 37 No. 7,
pp. 818-838.
Schillinger, P., Bürger, M. and Dimarogonas, D.V. (2019),
“Hierarchical LTL-Task MDPs for Multi-Agent
coordination through auctioning and learning”,The
International Journal of Robotics Research,
Schneider, E., Sklar, E.I. and Parsons, S. (2017), “Mechanism
selection for multi-robot task allocation”,in: Annual
Conference Towards Autonomous Robotic Systems,Springer.
Semwal, T., Jha, S.S. and Nair, S.B. (2017), “On ordering
multi-robot task executions within a cyber physical system”,
ACM Transactions on Autonomous and Adaptive Systems
(Systems), Vol. 12 No. 4, pp.1-27.
Shenoy, M.V. and Anupama, K. (2017), “DTTA-Distributed,
time-division multiple access based task allocation
framework for swarm robots”,Defence Science Journal,
Vol. 67 No. 3, p. 316.
Su, X., Wang, Y., Jia, X., Guo, L. and Ding, Z. (2018), “Two
innovative coalition formation models for dynamic task
allocation in disaster rescues”,Journal of Systems Science and
Systems Engineering, Vol. 27 No. 2, pp. 215-230.
Sung, Y., Budhiraja, A.K., Williams, R.K. and Tokekar, P.
(2018), “Distributed simultaneous action and target
assignment for multi-robot multi-target tracking”,in: 2018
IEEE International conference on robotics and automation
(ICRA),IEEE.
Talebpour, Z. and Martinoli, A. (2018), “Risk-based human-
aware multi-robot coordination in dynamic environments
shared with humans”,in: 2018 IEEE/RSJ International
Conference on Intelligent Robots and Systems (IROS),IEEE.
Tolmidis, A.T. and Petrou, L. (2013), “Multi-objective
optimization for dynamic task allocation in a multi-robot
system”,Engineering Applications of Artificial Intelligence,
Vol. 26 Nos 5/6, pp. 1458-1468.
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
941
Tsang, K.F.E., Ni, Y., Wong, C.F.R. and Shi, L. (2018), “A
novel warehouse multi-robot automation system with semi-
complete and computationally efficient path planning and
adaptive genetic task allocation algorithms”,in: 2018 15th
International Conference on Control, Automation, Robotics and
Vision (ICARCV),IEEE.
Turner, J. (2018), “Distributed task allocation optimisation
techniques”,in: Proceedings of the 17th international conference
on autonomous agents and multiagent systems.
Turner, J., Meng, Q., Schaefer, G., Whitbrook, A. and
Soltoggio, A. (2017), “Distributed task rescheduling with
time constraints for the optimization of total task allocations
in a multirobot system”,IEEE Transactions on Cybernetics,
Vol. 48 No. 9, pp. 2583-2597.
Wang, J., Gu, Y. and Li, X. (2012), “Multi-robot task
allocation based on ant colony algorithm”,Journal of
Computers, Vol. 7 No. 9, pp. 2160-2167.
Wang, X., Rui, F. and Hu, H. (2018), “Task allocation policy
for UGV systems using colored petri nets”,in: 2018 Annual
American Control Conference (ACC),IEEE.
Wawerla, J. and Vaughan, R.T. (2010), “A fast and frugal
method for team-task allocation in a multi-robot
transportation system”,in: 2010 IEEE International
Conference on Robotics and Automation,IEEE.
Wei, C., Hindriks, K.V. and Jonker, C.M. (2016), “Dynamic
task allocation for multi-robot search and retrieval tasks”,
Applied Intelligence, Vol. 45 No. 2, pp. 383-401.
Whitbrook, A., Meng, Q. and Chung, P.W. (2019),
“Addressing robustness in time-critical, distributed, task
allocation algorithms”,Applied Intelligence, Vol. 49 No. 1,
pp. 1-15.
Woosley, B. and Dasgupta, P. (2018), “Integrated real-time
task and motion planning for multiple robots under path and
communication uncertainties”,Robotica, Vol. 36 No. 3,
pp. 353-373.
Wu, D., Zeng, G., Meng, L., Zhou, W. and Li, L. (2017),
“Gini coefficient-based task allocation for multi-robot
systems with limited energy resources”,IEEE/CAA Journal of
Automatica Sinica, Vol. 5 No. 1, pp. 155-168.
Xie, B., Gu, X., Chen, J. and Shen, L. (2018), “A multi-
responsibility–oriented coalition formation framework for
dynamic task allocation in mobile–distributed multi-agent
systems”,International Journal of Advanced Robotic Systems,
Vol. 15 No. 6, pp. 1729881418813037.
Yi, X., Zhu, A., Yang, S.X. and Luo, C. (2016), “Abio-
inspired approach to task assignment of swarm robots in 3-D
dynamic environments”,IEEE Transactions on Cybernetics,
Vol. 47 No. 4, pp. 974-983.
Zhang, K., Collins, E.G., Jr. and Shi, D. (2012), “Centralized
and distributed task allocation in multi-robot teams via a
stochastic clustering auction”,ACM Transactions on
Autonomous and Adaptive Systems (Systems), Vol. 7 No. 2,
pp. 1-22.
Zhao, W., Meng, Q. and Chung, P.W. (2015), “A heuristic
distributed task allocation method for multivehicle multitask
problems and its application to search and rescue scenario”,
IEEE Transactions on Cybernetics, Vol. 46 No. 4, pp. 902-915.
Zhou, X., Wang, H., Ding, B., Hu, T. and Shang, S. (2019),
“Balanced connected task allocations for multi-robot
systems: an exact flow-based integer program and an
approximate tree-based genetic algorithm”,Expert Systems
with Applications, Vol. 116, pp. 10-20.
Zhu, Z., Tang, B. and Yuan, J. (2017), “Multirobot task
allocation based on an improved particle swarm optimization
approach”,International Journal of Advanced Robotic Systems,
Vol. 14 No. 3, p. 1729881417710312.
Zhu, D., Cao, X., Sun, B. and Luo, C. (2017), “Biologically
inspired self-organizing map applied to task assignment and
path planning of an AUV system”,IEEE Transactions on
Cognitive and Developmental Systems, Vol. 10 No. 2,
pp. 304-313.
Corresponding author
Mukund Nilakantan Janardhanan can be contacted at:
mukund.janardhanan@leicester.ac.uk
For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com
Task allocation strategies
Seenu N. et al.
Industrial Robot: the international journal of robotics research and application
Volume 47 · Number 6 · 2020 · 929–942
942
