Journal of Control, Automation and Electrical Systems
https://doi.org/10.1007/s40313-018-0390-y
Humanoid Robot Framework for Research on Cognitive Robotics
Danilo H. Perico
1
· Thiago P. D. Homem
1,2
· Aislan C. Almeida
1
· Isaac J. Silva
1
· Claudio O. Vilão Jr.
1
·
Vinicius N. Ferreira
1
· Reinaldo A. C. Bianchi
1
Received: 29 November 2017 / Revised: 26 March 2018 / Accepted: 1 May 2018
© Brazilian Society for Automatics–SBA 2018
Abstract
This paper presents a humanoid robot framework, composed of a simulator and a telemetry interface. The framework is based
on the Cross Architecture, and it is developed aiming for the RoboCup Soccer Humanoid League domain. A simulator is an
important tool for testing cognitive algorithms without handling issues of real robots; furthermore, a simulator is extremely
useful for allowing reproducibility of any developed algorithm, even if there is no robot available. The proposed simulator
allows an easy transfer of the algorithms developed in the simulator to real robots, as long as it uses the Cross Architecture as
its software architecture. Then, in order to evaluate the cognitive algorithms in real robots, a telemetry interface is proposed.
From this interface, it is possible to monitor any variable in the robot’s shared memory. The framework is open source and has
low computational cost. Experiments were conducted in order to analyze both, simulator and telemetry interface. Experiments
performed with the simulator aim to validate the high-level strategy development and the portability to a real robot, while
experiments with telemetry interface aim to evaluate the robot behavior using, as input, the information received from the
shared memory passed by all processes. The results show that the simulator can be used to test and develop new algorithms,
while the telemetry can be used to monitor the robot, thus validating the framework for this domain.
Keywords Humanoid robot framework · Cognitive robotics · Robot simulator and telemetry
1 Introduction
The humanoid robot framework proposed on this work is
based on the Cross Architecture (Perico et al. 2014a, b). The
Cross Architecture is a software architecture composed by
independent modules, that was proposed in order to allow a
humanoid robot, equipped with a computer, to perform sev-
eral concurrent tasks, including servomotor control, decision,
localization, vision and so on.
Several modules of the Cross Architecture are dedicated
to the development of cognitive algorithms, i.e., algorithms
aimed at building intelligent robots, where the input–output
behavior of the robot matches corresponding human behav-
ior (Russell and Norvig 2010). Nevertheless, one of the
main problems of developing cognitive algorithms in real
B Danilo H. Perico
dperico@fei.edu.br
Thiago P. D. Homem
thiagohomem@ifsp.edu.br
1
Centro Universitário FEI, São Bernardo do Campo, SP, Brazil
2
Instituto Federal de São Paulo, São Paulo, SP, Brazil
humanoid robots is the physical issues, such as backlashes,
broken parts, bad connections. Another problem of devel-
oping cognitive algorithms in unique humanoid robots, i.e.,
those ones that are not standard, is the lack of reproducibility
of any proposed code.
Thereby, a simulator becomes an important tool for debug-
ging and testing high-level algorithms without the need of
dealing with real robot problems. Furthermore, a simulator
allows the reproducibility of any proposed algorithm, even if
there is no real robot available. The usage of simulators also
allows the performance of a large number of experiments,
without the risk of damaging robots.
There are already several kinds of robot simulators avail-
able, such as Gazebo (2016), Webots (2016), Virtual robot
experimentation platform (2016) and RoboCup Soccer 3D
(2017). However, those 3D simulators often have high com-
putational cost and they need to have the specific desired
robot model, which is not always available by default. In
the existing 2D simulators, such as the RoboCup Soccer 2D
(2017), some of the difficulties arise from the integration of
the Cross Architecture. Beyond that, it is not an easy task to
123