IEEE Robotics & Automation Magazine - September 2019 - 87

To mitigate this problem, we have also added a complete set
of instructions with the various packages to install on a native
Linux system to enable reproducibility even without the VirtualBox image. Nevertheless, because of continuous updates
in software packages, as well as discontinuations, this second
approach may not be stable in the long run. To the best of our
knowledge, this is the first example of a fully reproducible
study of exploration algorithms.
Conclusions and Future Work
In this article, we studied an exploration task with temporal
constraints. The objective for the robot is to enter an unexplored area and reach a target location within a given temporal deadline. Our system incrementally builds a spatial
model called a semantic topological-oriented map. The
model enhances classic topological maps by adding semantic labels (corridor/room) and relationships of the type "to
the right of " and "to the left of." Key to our approach is the
assumption that walls in the environment are aligned along
orthogonal directions. This assumption is very common in
most buildings.
The proposed spatial model has been coupled with our
recently proposed planner based on CMDPs; in addition to
this, we proposed and analyzed five different exploration
techniques that exploit the proposed model. Through thousands of simulated runs, it appears that our method, dubbed
topological frontier with normalized distances, works best. In
their current formulation, methods on semantic information
are slightly less competitive, although it should be acknowledged that only very limited semantic information was used.
Finally, we have provided all of the resources to ensure that a
third party can fully reproduce our results-a first in the area
of reproducible robotics. With respect to this aspect, we
believe it will be important for the robotics community to
reach a consensus on the tools and infrastructure needed to
disseminate reproducible research.
References
[1] B. Yamauchi, "A frontier-based approach for autonomous exploration," in Proc. IEEE Int. Symp. Computational Intelligence Robotics and
Automation, Monterey, CA, 1997, pp. 146-151.
[2] B. Kuipers, "The spatial semantic hierarchy," Artif. Intell., vol. 119,
no. 1-2, pp. 191-233, 2000.
[3] E. Vidal, J. D. Hern and K. Isteniˇ, "Optimized environment exploration for autonomous underwater vehicles," in Proc. IEEE Int. Conf.
Robotics and Automation, 2018, pp. 6409-6416.
[4] T. Dang, C. Papachristos and A. Kostas, "Visual saliency-aware receding horizon autonomous exploration with application to aerial robotics,"
in Proc. IEEE Int. Conf. Robotics and Automation, 2018, pp. 2526-2533.
[5] D. Zhu, T. Li, D. Ho, C. Wang and M. Q. Meng, "Deep reinforcement
learning supervised autonomous exploration in office environments,"
in Proc. IEEE Int. Conf. Robotics and Automation, 2018, pp. 7548-7555.
[6] K. Antonios and G. Ioannis, "Semantic mapping for mobile robotics
tasks: A survey," Robot. Auton. Syst., no. 66, pp. 86-103, Apr. 2015.
[7] S. Garg, N. Suenderhauf and M. Milford, "Don't look back: Robustifying place categorization for viewpoint- and condition-invariant

place recognition," in Proc. IEEE Int. Conf. Robotics and Automation,
2018, pp. 3645-3652.
[8] M. Durner, M. Brucker, A. Wendt, P. Jensfelt, K. O. Arras and R.
Triebel, "Semantic labeling of indoor environments from 3D RGB
maps," in Proc. IEEE Int. Conf. Robotics and Automation, 2018,
pp. 1871-1878.
[9] L. F. Posada, A. Velasquez-Lopez, F. Hoffmann, and T. Bertram,
"Semantic mapping with omnidirectional vision," in Proc. IEEE Int.
Conf. Robotics and Automation, 2018, pp. 1901-1907.
[10] A. Quattrini Li, R. Cipolleschi, M. Giusto, and F. Amigoni, "A
semantically-informed multirobot system for exploration of relevant
areas in search and rescue settings," Auton. Robots, vol. 40, no. 4,
pp. 581-597, 2016.
[11] M. C. Gombolay, R. J. Wilcox, and J. A. Shah, "Fast scheduling of
robot teams performing tasks with temporospatial constraints," IEEE
Trans. Robot., vol. 1, no. 34, pp. 220-239, 2018.
[12] J. Guerrero and G. Oliver, "Swarm-like methodologies for executing tasks with deadlines," J. Intell. Robot. Syst., vol. 68, no. 1, pp. 3-19,
2012.
[13] Y.-L. Chow, M. Pavone, B. Sadler, and S. Carpin, "Trading safety
versus performance: Rapid deployment of robotic swarms with robust
performance constraints," ASME J. Dynam. Syst., Measurement Control, vol. 3, no. 137, pp. 031005-031005-11, 2015. doi: 10.1115/1.4028117.
[14] S. Carpin and S. Feyzabadi, "Planning using hierarchical constrained Markov decision processes," Auton. Robots, vol. 8, no. 41,
pp. 1589-1607, 2017.
[15] J. L. Susa Rincon, P. Tokekar, V. Kumar, and S. Carpin, "Rapid
deployment of mobile robots under temporal, performance, perception, and resource constraints," IEEE Robot. Autom. Lett., vol. 4, no. 2,
pp. 2016-2023, 2017.
[16] E. Garcia-Fidalgo and A. Ortiz, "Vision-based topological mapping
and localization methods: A survey," Robot. Auton. Syst., vol. 64,
pp. 1-20, Feb. 2015.
[17] E. Altman, "Constrained Markov decision processes with total
cost criteria: Occupation measures and primal LP," Math. Methods
Oper. Res., vol. 43, no. 1, pp. 45-72, 1996.
[18] D. P. Bertsekas, Dynamic Programming and Optimal Control,
vol. 1-2. Belmont, MA: Athena Scientific, 2005.
[19] S. Carpin, N. Basilico, D. Burch, T. Chung, and M. Kölsch, "Variable
resolution search with quadrotors: Theory and practice," J. Field
Robot., vol. 5, no. 301, pp. 685-701, 2013.
[20] R. Peng, "A simple explanation for the replication crisis in science,"
Simply Statistics, 2016. [Online]. Available: https://simplystatistics
.org/2016/08/24/replication-crisis/
[21] F. Bonsignorio, J. Hallam, and A. del Pobil, "GEM guidelines," Euron
GEM SIG, Genova, Italy, 2008. [Online]. Available: http://www.heron
robots.com/EuronGEMSig/downloads/GemSigGuidelinesBeta.pdf

Jose Luis Susa Rincon, Department of Computer Science and
Engineering, University of California, Merced. Email: jsusa
rincon@ucmerced.edu.
Stefano Carpin, Department of Computer Science and
Engineering, University of California, Merced. Email:
carpin@ieee.org.

SEPTEMBER 2019

*

IEEE ROBOTICS & AUTOMATION MAGAZINE

*

87
https://simplystatistics.org/2016/08/24/replication-crisis/ https://simplystatistics.org/2016/08/24/replication-crisis/ http://www.heronrobots.com/EuronGEMSig/downloads/GemSigGuidelinesBeta.pdf http://www.heronrobots.com/EuronGEMSig/downloads/GemSigGuidelinesBeta.pdf
IEEE Robotics & Automation Magazine - September 2019

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - September 2019
