Page 34 - Read Online

Page 127 Bernardo et al. Intell Robot 2021;1(2):116-30 I http://dx.doi.org/10.20517/ir.2021.10

Table 1. Example: generated plan
Global Time Action and respective objects used Time of action
0.000 (undock robot_base dock) [5.000]
5.001 (localise robot_base) [10.000]
15.002 (open robotiq robot_base) [2.000]
17.002 (move_base dock LivingRoom robot_base) [5.000]
22.003 (move_ur3 p0 p3 LivingRoom robot_base) [5.000]
27.003 (pick obj1 p3 robotiq robot_base LivingRoom) [2.000]
29.003 (move_base LivingRoom BedRoom robot_base) [5.000]
34.003 (move_ur3 p3 p2 BedRoom robot_base) [5.000]
39.003 (drop obj1 p2 robotiq robot_base BedRoom) [2.000]
41.003 (dock robot_base dock) [5.000]

(b) Execution of the action move_base and consult
MongoDB database

(a) Execution of the action undock

Figure 8. Execution of a plan by the robotic agent (AMMR).

requests that the agent should leave the dock, the action responsible for the task, /rosplan_interface_undock, is
triggered. This action in turn communicates with the action /move_base that communicates with the actuators
(motors) to move the robot. Figure 8b shows that, when the robot is requested to move to a certain room,
/rosplan_interface_move_base is activated, which in turn consults the MongoDB database to know where the
room is located in the map. Then, the robot moves in the environment, after calling /move_base action.

5. CONCLUSION AND FUTURE WORK
The use of ontologies has become a great solution, and one of the paths to follow in the future to make domain
knowledge explicit and eliminate ambiguities, allow machines to reason, and facilitate knowledge sharing be-
tween machines and humans. In this work, a structured ontology is presented to be used by robotic agents
in order to assist them in their deliberation tasks (interaction with the environment and robot movement).
It is imperative to endow robotic agents with semantic knowledge. Several approaches in the literature show
advantages in systems using databases such as MongoDB, pointing to their speed of response compared to
ontology-based systems. This paper introduces a framework that combines both, in terms of concepts and
their implementation in real robotic systems.