Aerospace and Electronic Systems - March 2019 - 53

Ropero et al.
Table 1.

ARIES Results Executing the Nominal Scenario Consisting of Acquiring Six Pictures in a Mars-Like Environment

Nominal

# comms

# Charging
Stops

Deliberation
Time (s)

Simulation
Time (s)

Distance
UGV (m)

Distance
UAV (m)

25.2

459.9

138.6

125.5

Table 2.

Results of the Three Missions Consisting of Acquiring Different Targets Number and UAV Battery Capacities in a Mars-Like
Environment
# comms

# Charging
Stops

Deliberation
Time (s)

Simulation
Time (s)

Distance
UGV (m)

Distance
UAV (m)

Low-Capacity 1

27.8

469.6

146.8

71.4

Nominal 2

128

40.3

668.1

152.1

168.2

Low-Capacity 2

146

44.2

761.5

158.2

52.4

tokens exchanged), the charging stops performed by the
UAV to reach the targets, the deliberation time to obtain a
plan, the simulation time taken by V-REP to perform the
simulation, and the distances traveled by the UGV and
UAV. We can observe that ARIES performs 64 communications between the leader and the follower to accomplish
the mission. As we can appreciate in the partial plan shown
in Figure 12, eight actions execution (blue boxes) in the
Follower Agent require eight effect tokens, so the complete
Nominal scenario with 32 actions requires a total of 64
communications. The UAV's high capacity battery allows
the hybrid UGV-UAV system to reach the six targets by
computing only four secure locations [in green in Figure 10
(a)]. Also, the UAV needs to perform only five charging
stops to complete the exploration without having energy
issues. Note the UP2TA planner aims to minimize the
UGV's distance traveled because the UGVs are used to be
slower robotic systems than the UAVs, which explains that
the UGV's distance traveled is less than the UAV's distance traveled.

COMPUTATIONAL RESULTS FOR DERIVED SCENARIOS
This section presents overall computational results executing ARIES over scenarios derived from the presented mission proposal. In the following, the tested scenarios are
described.
Low-Capacity 1. Starting at the home location, the
objective is to take six pictures. The UAV has a low battery capacity, so it requires the UGV to travel more distance in order to deploy the UAV closer to the targets.
MARCH 2019

Therefore, it is a slightly complex scenario than the Nominal, and so, the planner takes more time to compute a plan.
Nominal 2. Starting at the home location, the objective
is to take 12 pictures. In this scenario, the UAV has a high
capacity battery. This scenario is a more complex than the
previous one from the planning perspective as it is
required to plan more targets.
Low-Capacity 2. It is similar to the Nominal 2, but the
UAV has a lower battery capacity. Thus, it represents a
combination of the previous scenarios where the UGV
needs to travel a longer distance to allow the UAV to
reach those targets, while the planner has to deal also with
a high number of targets (12 pictures).
The results of executing such scenarios are presented in
Table 2. First, we can observe that the communications and the
charging stops are increased either by the task number or the
battery capacity. Second, the deliberation and simulation times
increase due to the scenario complexity. Finally, the UGV's
distance traveled is higher in the low-capacity scenarios
because it is required to deploy the UAV closer to the targets
to allow it to reach the objectives, which also explains that the
UAV's distance traveled is quite lower on these scenarios.

CONCLUSION
This paper is focused in cooperative robotics research,
particularly for space exploration missions. To do this,
we proposed an autonomous controller developed to
deploy cooperative heterogeneous robot systems which
combine and coordinate their functionalities to achieve
scientific targets in challenging scenarios. Contrary to

IEEE A&E SYSTEMS MAGAZINE

Aerospace and Electronic Systems - March 2019

Table of Contents for the Digital Edition of Aerospace and Electronic Systems - March 2019