Victims Search, Identification, and Evacuation with Heterogeneous

Robot Networks for Search and Rescue

David Yanguas-Rojas1 , Gustavo A. Cardona 2 , Juan Ramirez-Rugeles 3 and Eduardo Mojica-Nava4

Abstract—This paper presents a control strategy for multiple can be explored faster with the assistance of the robots giving
heterogeneous robots oriented to the assistance in situations of precious time to the paramedics to attend the victims. The
search and rescue from two complementary perspectives, the use of the robots can reduce the number of victims in the
real time control and the discrete task allocation. The real time
control laws are in charge of the correct execution of the tasks situation since more lives will not be risked when searching
assigned by the reallocation algorithm respecting constraints due to the area can still be dangerous, so we can lain over them
in the connectivity range, the collision avoiding and the full to perform this tasks while the human resources are focused
completion of each task. The considered tasks are exploration in the paramedical attention and the evacuation of people in
of the mission environment, search and identification of victims, difficult medical conditions.
delivery of medical supplies to victims unable to move and
evacuation of victims capable to move. During the development of This opens the door to the study of which tasks can be
each task, the robots take routes optimized considering distance performed by the robots in search and rescue of victims. Tasks
metrics based on the breath-first search algorithm and an such as exploration, search, delivery of supplies, localization,
exploration algorithm for non-convex environments. In order rescue of survivors, among others, are the typical tasks studied
to assign the tasks to the robots thorough the mission, we for them. However there are multiple tasks different that can
present an optimized task reallocation algorithm focused on the
minimization of the time required to attend all the victims in be made for them such as evacuation of people or firefighting.
the mission environment. This algorithm allows to assign each There are many theories and research work applied to this
of the robots to an appropriate task considering that the robots topic of robotics that try to solve different challenges of
may differ in the capability of completing each task as well as the previously mentioned tasks, such as the research made
in their moving capabilities. This complete framework allows a in [4] that shows the need to use heterogeneous robotics
team of autonomous robots to bring a valuable assistance in a
search and rescue situation and its behavior is presented in a teams in search and rescue operations, in [5], it is presented
non-trivial simulation scenario. schemes of heterogeneous robots team formation, while the
Index Terms—Multi-Robots Systems, Rescue Robots, Robot work presented in [6] exposes an exploration algorithm using
Control heterogeneous robots. Other common approach it is the use
of swarm theory applied into exploration and recognition of
I. INTRODUCTION
large areas a short time, compared with the time employed by
The worst natural disasters and terrorist attacks of mankind a single robot for performing the same task as presented in
history have motivated multiple researchers around the world [7]. It is important to mention that the work presented in [8]
to study the rescue robotics area, for example in [3] it is and in [9] are a benchmark to our contribution on exploration
presented how robotics can impact on some events such as task with their use of Voronoi tessellation on homogeneous
the attacks of the World Trade Center. Unfortunately, most of networked robots and the use of artificial potential functions,
those disaster situations cannot be detected on time, causing respectively.
populations to be unable to evict the area, to be unable to In allocation tasks there are some other researchers that are
be ready for this kind of situations and to have problems working on this area, for example in [10] there are algorithms
avoiding a large number of human and animal casualties. presented for task allocation challenges in robotics search and
When a disaster occurs, it is really important to have a solid rescue operations, in the same way in [11] it is analyzed
contingency procedure as explained in the work presented the robots task allocation with cost uncertainties for different
in [2] which shows how the first hours and initial days are situations, and finally in [13] it is shown a control strategy
the most important in the rescue of survivors. Taking this employable for the coordination of multi-robots to different
into account the use of networked heterogeneous robots to kinds of tasks.
collaborate the rescuers in a disaster area can improve the The other sections that complete this paper are organized
chance of survive for the people trapped in it. Also the area as follows: Section 2 presents the exploration of the disaster
1 D. Yanguas-Rojas is from Faculty of Engineering, University National of environment, Section 3 the optimized task allocation and
Colombia Bogota, Colombia Section 4 simulation results.
2 G. Cardona is from Faculty of Engineering, University National of
Colombia Bogota, Colombia II. E XPLORATION OF THE D ISASTER E NVIRONMENT
3 J. Ramirez-Rugeles is from Faculty of Engineering, University National
of Colombia Bogota, Colombia The efficiency on exploration task is one of the most
4 E. Mojica-Nava is from Faculty of Engineering, University National of
important things in SAR (Search and Rescue) because the
Colombia Bogota, Colombia velocity and the way to do that, can define the survival of
2017 IEEE 3rd Colombian Conference on Automatic Control (CCAC) humans that be in distress or imminent danger. That is the
978-1-5386-0398-7/17/$31.00 2017
c IEEE
reason why one of the major aims on search and rescue area. During the mission, the network changes constantly
robotics topics is to find an optimized method to explore and since the robots are moving at different speeds as they are
map the disaster area due that frequently there may not be approaching to the exploration frontiers. This makes some
information available about the state of the environment after links useless and requires the creation of new links, so it
a disaster occurs in an urban area or rural area. The correct was necessary a network reconfiguration algorithm capable
development of this task allows to reduce the time and the of maintaining connectivity and flexibility. It is worth to
difficulty to recognize and localize all the victims as soon as mention that considering that sometimes the robots can reach
possible. The method contemplated in this work to exploration undesired equilibrium points where they reach the centroid
task is given by: of the Voronoi cell and the exploration task is not complete
yet, we use a perturbation signal which allows the robots to
A. Heterogeneous DisCoverage
escape from these equilibrium points respecting the effect of
In the present paper, we extend a previous work presented in the artificial potential keeping the network connectivity and
[1] where there are modifications to the DisCoverage algorithm the collision avoidance.
presented in [8] in order to adapt it to the heterogeneous case.
This frontier based algorithm tracks the weighted centroid of B. Modifications to the Exploration Algorithm
certain area (assigned to each robot using a Voronoi based One of the biggest changes in the algorithm is how the
tessellation) allowing the robots network to explore since the breadth first search is implemented, in the paper presented
weight function grows near to the map’s frontier. There, the before [1], there is used a 4 neighbors flooding, which means
tessellation employed considers the different moving speeds that the algorithm was comparing an initial cell’s value with
for the robots resulting in cells different from the traditional the cells above, under, right side and left side of the initial cell,
Voronoi Cells previously employed with homogeneous robots considering only movements in strictly horizontal and vertical
with the same speed. This happens because the criteria of directions, what can be consider as the representation of the
tessellation are not anymore the points that are closest to the Manhattan distance in a grid or when the grid tends to the
robot, but the points that the robot can reach the fastest. continuous case to the norm L1 . Now the algorithm is using a
Another consideration taken in the previous work is that, 8 neighbors flooding (8NF), allowing to not just going through
the disaster area is a non-convex space where the euclidean the directions mentioned before but also through the ones
distance direct calculation may be inaccurate. This makes located immediately on the diagonals (NE, NW, SW and SE).
necessary the implementation of another kind of distance The improvement of this change is that the distance is way
metric such as 4-neighbors flooding distance based on the more accurate and tends to be closer to the Euclidean distance.
Breadth first search presented in [14] or the use of a optimized In the algorithm implemented for flood distance, we changed
distance metric considering that the robot is capable to move the term that updates the value of the cell by comparing it with
on any direction how it is shown in Fig. 1 where the direct the current
calculus of the euclidean distance is not correct since the √ value plus 1 if it is in the 4 neighbors directions
or plus 2 if is in the diagonal directions (the corresponding
obstacles do not allow to follow a direct path. euclidean distance for step the movement). In the next image
we can realized the effect of this change (See Fig.2) where the
euclidean distance, the Manhattan distance and the presented
8NF are compared. It is worth to mention that an error can be
produced when the robot is calculating in different angles than
the ones represented by n45◦ , because the main calculus are
performed in those angles. The maximum error presented in
the calculus is when the robot sense in the angles represented
by n45◦ + 22.5◦

Fig. 1: Green dot: Actual position Red dot: Target position.

Red line: Euclidean distance. Blue line: Strictly optimized
distance in a non-convex environment.

An important issue boarded in the previous work that affects

the right performance of a mobile robots network is the Fig. 2: Green line: Manhattan distance. Red line: euclidean
connectivity maintaining and the collision avoidance of the distance. Blue line: equivalent to 8 breadth search.
robots. As presented in [9], artificial potential functions are
applied to keep the distance between robots within the limits In Fig.3 it is shown the comparison between the strictly
of the communication range, but with the possibility of doing optimal trajectory and the trajectories made by the theoretical
reconfigurations in order to extend the network over a bigger 4NF and 8NF, (3a and 3b). Then we have the strictly optimal
trajectory against the real trajectories of the 4NF and 8NF (3c Algorithm 1 Exploration Behavior
.
and 3d). The real trajectories are those where the robots also
1: function E XPLORATION (Robots, scene) Where Robots
apply the net virtual potential functions to avoid collisions.
We can appreciate how the robot manage to trace a trajectory - robots array, scene - matrix
that tends to the strictly optimal, when the robot applies the 2: for r ∈ Robots do
8NF and can still avoid the walls by the potential functions 3: Heterogeneous Voronoi Partition(Robots,Scene)
respecting the security ranges. (3d) 4: if dS ∈ Vr = {φ} then
5: Track Voronoi Centroid
6: NV F (r) = 0
7: else
8: N CF ← Find min NV F (rn) for rn ∈
Robot sN eighbors
9: Track N CF
10: NV F (r) = NV F (N CF ) + 1
11: end if
12: end for
13: end function
(a) Optimal convex distance vs (b) Optimal convex distance vs
4NF 8NF
III. O PTIMIZED TASK A LLOCATION
Throughout the exploration process there are multiple emer-
gent task that must be done by the rescue team. Some of
them could only be made by humans like the first aid and
paramedical tasks, however there are many kinds of tasks
that can be realized by the robots like the presented in
[12]. In this paper, we are going to consider mainly the
(c) Optimal convex distance vs (d) Optimal convex distance vs
victims identification, the victims evacuation and the delivery
Real 4NF Real 8NF of medical supplies. Those tasks must be allocated to the
robots prioritizing the minimization of the mission time that
Fig. 3: Comparison between convex optimal distance with is minimize the time that the robots take for accomplishing
4NF, 8NF, real 4NF and real 8NF the attention to all the victims and considering that the robots
may have different capacities for doing each task.
Another change that is worth to mention is that when new
tasks are assigned to the robots network or just to one of A. Task Definition
the robots, the control algorithm will perform a change in the Since the objective of the algorithms presented is to min-
exploration component, replacing the exploration component imize the time for accomplish the mission, all the task that
by the corresponding component of the new task. However will be developed by the robots must have an estimation of its
the net virtual potential will remain the same. This way the temporal cost associated in order to determine an optimized
control signal adopts the form ui = uti +upoti , where ui is the allocation. Those estimations are made as follows:
control signal for the i robot, uti corresponds to a proportional 1) Exploration: The first task that each robot must do is
controller that tracks the mission point of interest (Voronoi the exploration of the scene as mentioned before, however the
centroid, victim location or evacuation point) for the robot i cost value in time of this task is not trivial and the estimation
and upoti is the net artificial potential for the robot i. that we took is presented in (1) in such a way that in the early
The final important change made to the original algorithm stages of the mission the exploration takes a high priority and
is the way the robots r that do not have frontier dS within its in the final stages of the mission it may be even discarded if
own Voronoi cell Vr behave. In the previous work the Voronoi there are not more expected victims to be found.
centroid of the cell without frontier became its geometric cell
Am
and the robot will track it. However, that leads the robots to Cie = Ke (1)
stay in areas of low interest since there were already explored. Vi Nmv
Considering that, we changed this behavior for tracking the where Cie is the cost of the exploration task for the robot
neighbor N CF who has the frontier the closest (in terms of i, Ke is an arbitrary constant that allows to give priority to
nodes of the graph NN F (r)). This way if the robot has no the attention of the already found victims or to the finding
frontier within its cell it will go closer to the limits of the of the expected missing victims and allows the conversion of
exploration frontier and will help with the exploration process the result of the expression to seconds, Am is the area of
speeding up the whole process as is presented in the Algorithm the mission environment that has not been yet explored, Vi is
1. the speed of the robot i and Nmv is the expected number of
 V
missing victims to be found. The variable Nmv must be set by where Ci,j is the cost of the robot i for evacuate the victim j,
a upper bound of the number of people expected to be found Tr,Ev is the time required for the robot i to communicate to
in the disaster area and can be estimated considering the usual a victim to follow it, dj,SA is the distance between the victim
population density of the location or logistics information j and the safe area and Vj is the speed of the victim j.
available before the emergency situation.
2) Identification: Concerning to the victims we have con- B. Events and Task Allocation
sidered three types of them. The victims that are capable of During the start of the mission, all the robots are set
moving by themselves and can be evacuated, the victims that to explore the mission environment since the position of
are alive but cannot move or require human assistance and the victims is unknown so the allocation is trivial. When a
the casualties or false victims that cannot be helped. Once a victim is found or a task is concluded, the tasks available
victim is found the next task to be done is its classification and to be done change so the previous task allocation must be
since not necessarily all the robots can, there is the possibility reevaluated and since generally there will be more tasks than
that the robot that has detected the victim is not capable of robots it is necessary to evaluate which task allocation is
classify or requires to be nearer into the sensing range in order optimal in the current setup. In order to solve this allocation
to do the classification. The cost for this task is presented problem we utilize the extension of the Hungarian method
in (2) taking advantage of the distance matrix calculus with for rectangular matrix presented in [16] with the current cost
the 8 neighbors-flooding which allows the estimation of the matrix considering that all the robots can explore (if there is
minimum distance from each robot to the victim even if the space pending to be explored) and that if there are victims that
path to it is not linear. are not satisfactorily attended the robots that can attend them
will have the possibility. Note that if a robot is not capable
ID di,j − ri,ID of realizing determined tasks the correspondent cost will be
Ci,j = Ti,ID + (2)
Vi infinity and the Hungarian method will not assign that task to
where Ci,jID
is the cost of the robot i for identifying the any robot.
victim j, Ti,ID is the time required for the robot i for the 1) Event Triggered Reallocation: Since the mission condi-
identification of a victim (since they may have different types tions change in time we call the reallocation algorithm each
of sensors this time may vary), di,v is the distance between time that an important event occurs. Those events are the
the robot i and the victim j evaluated through the distance discovery of a potential victim, the identification of the type
matrix generated using the 8 neighbors-flooding, ri,ID is the of a victim, the successful deliver of a supply, the successful
minimum distance required for the robot i to identify a victim evacuation of a victim and the end of the exploration task
and Vi is the speed of the robot i. (there is no more space to be explored). Since those are the
3) Supplies Deployment: if the victim is alive but cannot uniques events considered that change the tasks costs table
be evacuated the robots may set a transmitter for marking and the Hungarian method guarantees the optimality of the
precisely its location for the human rescue teams or depending assignation, we can be certain that in every moment of the
of the situation may deliver first aid kits or gas masks among mission but the moments during the reallocation takes place
others. This task requires that the robot reaches the location of the task allocation will be optimal.
the victim and executes the delivery, this is presented in (3). 2) The Reallocation Propagation: In order to respond to the
changes in the mission environment the fastest as possible, the
Del di,j robot that detects the change is in charge of evaluating the task
Ci,j = Ti,Del + (3)
Vi reallocation and send the message to all the other members of
Del the team. Note that the message will be always delivered to all
where Ci,j is the cost of the robot i for delivering a supply
the robots of the team since the communication is guaranteed
to the victim j and Ti,Del is the time required for the robot i
by the artificial functions that preserve the existing links in the
to hand over the supply.
network and the network reconfiguration algorithm presented
4) Evacuation: if the victim is alive and can be evacuated
in the previous work.
the robot may guide it to a safe area so the robot must reach the
It is important to mention that the events that trigger the
location of the victim going to its own speed and then when
reallocation, in general are expected to not be simultaneous,
the victim is in range they must go to the safe area going at
however if there is the case when two robots present different
the smallest speed between the speed of the victim and the
task allocations based on two simultaneous events the com-
speed of the robot. This is summarized in (4) considering that
munication protocols will give an pseudo arbitrary order to
the speed of the victim depends on its state and the class of
the messages and the network will accomplish the consensus
the situation, for example the victim could move faster if is
based on the latest. Other possibility that could be considered
in perfect health, has good visibility and the air is clean but
is to run a simple exception algorithm which allows the restore
could move slower if is wounded, has poor visibility and/or
of the consensus of the mission information and generate the
there is smoke in the air.
correct optimal allocation. However in practical scenarios the
Ev di,j dj,SA final behavior is expected to be very similar and the first
Ci,j = Tr,Ev + + (4) approach was considered in this work.
Vi min(Vi , Vj )
 IV. S IMULATION R ESULTS
In Fig. 4 we present the map corresponding to a complete
mission scenario where the robots have to explore the scene
in the search of the potential victims, determine the state of
each one, accomplish the tasks corresponding to each victim
and return to a Home position. The simulation setup employed
is the same that in the earlier version of the last simulation
presented in [1].

Fig. 5: Exploration sequence of the robot network with the

Fig. 4: Scene and Victims Location for the Simulations exploration improvements

A. Exploration Improvement enhancing performance and reducing time in the exploration

The main change in the exploration algorithm was made task, while maintaining connectivity.
over the behavior of a robot which has finish the exploration
task due to presenting no frontier (δS) to explore and does B. Full Mission Simulation
not have any other task assigned, normally it will move In this simulation the initial position of the robots is the
towards the center of its Voronoi cell. This is quite useless, the same line configuration presented in the previous subsection,
network can be using the resources of this robot to enhance however in this case there were 10 victims randomly allocated
the performance, here is where the new algorithm enters, now in the right half of the environment and with random status
the command is plain and simple this, if the robot does not (4 victims that can move, 3 victims that cannot move and 2
present frontier to explore, search for the nearest robot in the casualties). It is worth to mention that the victims capable
graph network structure (Rni ), that has frontier, then move to to move have simple dynamics programmed that consider the
the exact location of that robot. As soon as the wanderer robot collision evasion, follow a robot if the robot calls for it and go
reach the followed robot, it will have frontier at its sight so to the top left corner (safe zone) once the robots takes then near
it is going to start exploring once again, this will reduce the enough. They were programmed with a speed of 5 km/h that
exploration time that would be obtain if we had kept the old is a typical speed for a person walking. In Fig. 6a we present a
algorithm. situation when during the exploration task there were 4 robots
In the Fig.5, which is the same presented in Fig.4 without found and identified however the task allocation algorithm has
victims, we can appreciate the entire process, first we have the determined most of the robots to continue exploring and two
top left corner, where the entire robot network is deploy, we of them were set to deliver supplies to the non-capable to move
call this home position, then they start the exploring task, each victims.
robot has its own velocity so, we can see in the top right corner In Fig. 6b we present the moment when the robots found and
how the network expands to cover the left side of the area, at identified the last of the victims and did not explore anymore.
the moment just the robot a the front has frontier, the rest has In that moment all the non-capable to move victims have
no frontier to go to, so they will follow the one with frontier, already received their medical supplies so 4 of the robots were
in the bottom left corner, we have two robots with frontier and allocated to the evacuation task and the other were set to return
then two robots, surrounded by a green dot line, we will call home since there were not more task pending to be done. Note
them wanderers, those robots are the ones asking for neighbors that since all the victims expected were found the map was
with frontier, in the final photograph, bottom right corner, we not fully explored (remembering that the number of victims
can see how almost the entire area has been discover, from the expected is set to be higher or equal to the real number of
two wanderer robots one went to the upper side of the area victims in the scene). Later in Fig. 6c we present the process
to reach the fastest robot and the other one stayed in range to of evacuation, in this moment the victims are being evacuated
keep connectivity and to help with the frontier in the bottom by the robots to the top left corner of the environment (the
side of the area, it is quite clear how the robots reach those safe zone). Note that all the robots that are evacuating victims
with frontier as soon as they do not have any more to explore, are close enough to their victims, this behavior is expected
 interesting that can be made is a prediction based approach that
do not consider just the actual state of the environment and
victims, but estimations of the future positions of the victims
and possible locations for the remaining victims. Finally an
important consideration that can be taken in the future work
is to consider tasks interdependencies such as in [17] in a way
that allows the robots to do tasks more complex or to evaluate
sequential tasks like exploration, identification and evacuation
in a non-independent way.
(a) First Victims Identified (b) End of Exploration Process
R EFERENCES
[1] D. Yanguas-Rojas and E. Mojica-Nava, “Exploration with Heterogeneous
Robots for Search and Rescue”,’ IFAC 2017 World Congress of the
International Federation of Automatic Control’, 9-14 July 2017.
[2] R. C. Huder, Disaster Operations and Decision Making (1). Hoboken,
US: Wiley, 2013.
[3] J. Casper y R. R. Murphy, “Human-robot interactions during the robot-
assisted urban search and rescue response at the World Trade Center”,
IEEE Trans. Syst. Man Cybern. Part B Cybern., vol. 33, nm. 3, pp. 367-
385, jun. 2003.
[4] R. Ventura y P. U. Lima, “Search and Rescue Robots: The Civil
(c) Evacuation of capable victims (d) End of mission Protection Teams of the Future”, en 2012 Third International Conference
on Emerging Security Technologies (EST), 2012, pp. 12-19.
Fig. 6: Victims detection, classification and attention - The [5] Gunn, T., and Anderson, J. (2015). Dynamic heterogeneous team forma-
Blue Dots correspond to victims capable to move, the green tion for robotic urban search and rescue. Journal of Computer and System
ones correspond to the victims that are unable to move and Sciences, 81(3), 553-567.
[6] Chand, P., and Carnegie, D. A. (2013). Mapping and exploration in a
the yellow ones to the casualties found hierarchical heterogeneous multi-robot system using limited capability
robots. Robotics and autonomous Systems, 61(6), 565-579.
[7] Leon J., Cardona G.A., Botello A., Caldern J.M. (2017) Robot Swarms
and will be hold even if the victim’s speed and robot’s speed Theory Applicable to Seek and Rescue Operation. In: Madureira A.,
Abraham A., Gamboa D., Novais P. (eds) Intelligent Systems Design
differ (actually in most cases this is true). Finally in Fig. 6d and Applications. ISDA 2016. Advances in Intelligent Systems and
we show the final state of the mission, here the robots are set Computing, vol 557. Springer, Cham.
to an arbitrary formation, the capable victims were allocated, [8] Haumann, D., Willert, V., and Listmann, K. D. (2013). DisCoverage:
from coverage to distributed multi-robot exploration. IFAC Proceedings
the medical supplies were delivered and all the victims and Volumes, 46(27), 328-335.
casualties locations were recorded in the robots memories. [9] Zavlanos, M. M., Egerstedt, M. B., and Pappas, G. J. (2011). Graph-
theoretic connectivity control of mobile robot networks. Proceedings of
V. CONCLUSIONS AND FUTURE WORK the IEEE, 99(9), 1525-1540.
[10] Turner, J., Meng, Q., and Schaefer, G. (2015, May). Increasing allocated
In general, as it was expected, the desired behaviors of the tasks with a time minimization algorithm for a search and rescue scenario.
previous work is kept. The network connectivity was held; the In Robotics and Automation (ICRA), 2015 IEEE International Conference
collisions between robots, victims and obstacles were avoided on (pp. 3401-3407). IEEE.
[11] Nam, C., and Shell, D. A. (2015, May). When to do your own thing:
(note that the victim-victim collision was not considered since Analysis of cost uncertainties in multi-robot task allocation at run-time. In
in the real world the people will not collide themselves) and Robotics and Automation (ICRA), 2015 IEEE International Conference
the exploration process was successfully made. Regarding to on (pp. 1249-1254). IEEE.
[12] Pellenz, Johannes and Jacoff, Adam and Kimura, Tetsuya and Mi-
the exploration, we aimed to reduce the execution time of the hankhah, Ehsan and Sheh, Raymond and Suthakorn, Jackrit (2013).
network to explore an area and we succeeded. When the robots ’RoboCup rescue robot league’ in Lecture Notes in Artificial Intelligence
do not have frontier at sight they follow their neighbor that is (Subseries of Lecture Notes in Computer Science) (pp. 673685). Elsevier
Verlag.
the closest to the frontier speeding up the whole exploration [13] Lee, S. G., Diaz-Mercado, Y., and Egerstedt, M. (2015). Multirobot
process. In particular the employ of the 8NF metric made the control using time-varying density functions. IEEE Transactions on
distance calculus closer to the real distance in the non-convex Robotics, 31(2), 489-493.
[14] Skiena, S. S. (2008). Sorting and Searching. The Algorithm Design
domain and allowed the determination of optimized routes for Manual, 103-144.
the execution of most tasks. Note that the artificial potential [15] Zhang, B., Peng, J. (2013). Uncertain programming model for uncertain
functions in charge of the connectivity maintenance and the optimal assignment problem. In Applied Mathematical Modelling (pp.
64586468). Elsevier.
collision evasion change the real routes followed. [16] F. Burgeois and J.-C. Lasalle (1971). An extension of the Munkres
Concerning to the task definition and event based realloca- algorithm for the assignment problem to rectangular matrices. Commu-
tion proposed, they showed to be effective in the simulation nications of the ACM, 142302-806, 1971.
[17] Brutschy, A., Pini, G., Pinciroli, C. et al. Self-organized task allocation
environment, since they allowed the mission completion fully to sequentially interdependent tasks in swarm robotics In Autonomous
autonomous without any human assistance. It is important to Agents and Multi-Agent Systems (2014) 28: 101. doi:10.1007/s10458-
mention that the tasks cost definition can be improved to be 012-9212-y
more accurate with the real routes and that would be an in-
teresting next step to be followed. Another important upgrade