\titlecapThe Swarmathon: An Autonomous Swarm Robotics Competition

\titlecapThe Swarmathon: An Autonomous Swarm Robotics Competition

Sarah M. Ackerman, G. Matthew Fricke, Joshua P. Hecker, Kastro M. Hamed, Samantha R. Fowler,
Antonio D. Griego, Jarett C. Jones, J. Jake Nichol, Kurt W. Leucht, and Melanie E. Moses
Computer Science Department, The University of New Mexico, Albuquerque, NM, USA, Education and Interdisciplinary Studies, Florida Institute of Technology, Melbourne, FL, USA, NASA Kennedy Space Center, FL, USA, Department of Biology, The University of New Mexico, Santa Fe Institute, Santa Fe, NM, USA, This work was supported by NASA \acMUREP #NNX15AM14A and a generous donation from Google. We thank Theresa Martinez, \acMUREP STEM Engagement Manager; Paul Secor and Mary Baker of Secor Strategies, LLC; Kate Cunningham and Beatriz Palacios Abad of UNM ADVANCE; Elizabeth Esterly and the students of the Moses Biological Computation Lab, Vanessa Svihla, Aeron Haynie, NASA volunteers, UNM support staff, KSC Visitor Complex, and the Swarmathon students and faculty mentors.Correspondence: sherbet@unm.edu
Abstract

The Swarmathon is a swarm robotics programming challenge that engages college students from minority-serving institutions in NASA’s Journey to Mars. Teams compete by programming a group of robots to search for, pick up, and drop off resources in a collection zone. The Swarmathon produces prototypes for robot swarms that would collect resources on the surface of Mars. Robots operate completely autonomously with no global map, and each team’s algorithm must be sufficiently flexible to effectively find resources from a variety of unknown distributions. The Swarmathon includes Physical and Virtual Competitions. Physical competitors test their algorithms on robots they build at their schools; they then upload their code to run autonomously on identical robots during the three day competition in an outdoor arena at Kennedy Space Center. Virtual competitors complete an identical challenge in simulation. Participants mentor local teams to compete in a separate High School Division. In the first 2 years, over 1,100 students participated. 63% of students were from underrepresented ethnic and racial groups. Participants had significant gains in both interest and core robotic competencies that were equivalent across gender and racial groups, suggesting that the Swarmathon is effectively educating a diverse population of future roboticists.

\setremarkmarkup

(#2) \definechangesauthor[name=Matthew Fricke, color=darkgreen]GMF \definechangesauthor[name=Melanie Moses, color=orange]MEM \definechangesauthor[name=Sarah Ackerman, color=blue]SMA \Addlcwordsa an the at by for in of on to up and as but or nor so yet vs with per \DeclareAcronymISRU short = ISRU, long = in-situ resource utilization, \DeclareAcronymNASA short = NASA, long = National Aeronautics and Space Administration, \DeclareAcronymROS short = ROS, long = the robot operating system, cite = Quigley2009 \DeclareAcronymCS short = CS, long = computer science, \DeclareAcronymGPS short = GPS, long = global positioning system, \DeclareAcronymIMU short = IMU, long = inertial measurement unit, \DeclareAcronymPID short = PID, long = proportional-integral-derivative, \DeclareAcronymMSI short = MSI, long = minority-serving institution, \DeclareAcronymCPF short = CPF, long = central-place foraging, \DeclareAcronymMUREP short = MUREP, long = Minority University and Research Education Program, \DeclareAcronymSTEM short = STEM, long = science, technology, engineering, and mathematics, \DeclareAcronymABET short = ABET, long = Accreditation Board for Engineering and Technology, \DeclareAcronymEKF short = EKF, long = extended Kalman filter, cite=Moore2014 \DeclareAcronymGUI short = GUI, long = graphical user interface, \DeclareAcronymSIPI short = SIPI, long = Southwestern Indian Polytechnic Institute, \DeclareAcronymUNM short = UNM, long = the University of New Mexico,

I Introduction

I-a \titlecapThe Journey to Mars

The \acNASA Journey to Mars program has set the ambitious goal of sending manned missions to Mars by the 2030s.111https://www.nasa.gov/content/nasas-journey-to-mars These surface missions may last months or years, and because transporting sufficient materials from Earth is not practical, astronauts will need to utilize the resources already available on Mars. For example, small pockets of ice can be melted for water and converted to oxygen and hydrogen for fuel. This approach is known as \acISRU.

I-B \titlecapMeeting the Challenge with Robot Swarms

Autonomous robot swarms could locate, collect, and store resources in advance of human arrival. A swarm of small, inexpensive robots acting autonomously provides several advantages over a few large, expensive, manually controlled rovers [1, 2] because they are robust, scalable and able to explore unmapped environments [3].

\ac

ISRU is closely related to \acCPF, in which resources are transported to a central collection zone. Inspired by the success of social insects gathering resources for their colonies, an early study of robot foraging computationally modeled a swarm collecting rock samples on a distant planet [4]. \acCPF is a key robot swarm application [5, 6], and research continues to improve algorithms and engineering approaches for foraging swarms [7, 8].

Our team has designed and built a swarm of foraging robots called Swarmies. Swarmies are rugged enough to operate outdoors for hours, and feature sensors and grippers that enable them to complete CPF tasks. We also developed custom software packages and a \acGUI, enabling rapid development and testing of CPF algorithms in simulation and physical hardware.

Swarm foraging provides an ideal educational challenge because successful strategies require foundational robotics skills such as grasping, localization, navigation, exploration, and decentralized communication and coordination [9].

I-C The Swarmathon Competition

In the Swarmathon annual competition, teams of college students develop algorithms for autonomous swarm foraging. Teams in the Physical Competition program groups of 3 - 6 Swarmies to collect the most of resources placed in outdoor arenas. The Virtual Competition runs in a Gazebo simulation of the same arenas and robots. Robots must search for, collect, and return resources to a central collection zone completely autonomously. The competition runs in a series of rounds, and in each round resources are placed in different distributions (i.e., uniformly at random, or in various sizes and shapes of clusters), and the same code is run in each round. Score is determined by the number of resources in the collection zone at the end of the round (resources pushed out of the collection zone by robots do not count toward the score). Robots have limited view and no global map, making it difficult to find resources. In a third High School Competition, younger students are mentored by Swarmathon college students to complete a similar challenge in a simplified simulation. Winners of the Virtual competition advance to the Physical competition the following year.

I-D \titlecap Diversifying the STEM/Robotics Pipeline

Tackling \acNASA’s real-world \acISRU problem provides an opportunity to engage students in robotics and \acCS, helping meet the growing need for \acSTEM talent in the workforce. While women and minorities account for a growing proportion of US college graduates and US workers, they are underrepresented among \acSTEM college graduates and professionals, especially in \acCS and engineering [10, 11, 12, 13]. Their inclusion in STEM helps to address an anticipated shortfall in these fields over the next few decades [12, 13]. Additionally “workers from a variety of backgrounds enhance the quality of science insofar as they are likely to bring a variety of new perspectives…in terms of both research and application” [11]. For these reasons, the Swarmathon recruits competitors from 2- and 4-year colleges and universities that are \acpMSI including Historically Black Colleges and Universities (HBCUs), Hispanic-Serving Institutions (HSIs), and Tribal Colleges and Universities (TCUs). This provides an educational opportunity that may not otherwise exist for participating students:

“The Swarmathon is an opportunity for students to work on a real-life engineering problem that’s interdisciplinary and hard. They know that I don’t know the answer which makes it fun for all of us. I teach at a small community college and it simply wouldn’t be possible for us do this level of work from scratch.” -Swarmathon Faculty Mentor

The Swarmathon has engaged a diverse student population with 81% of students from minority groups (including Asian students) and 63% of the 818 college participants identifying as belonging to an underrepresented racial or ethnic group (Black/African American, American Indian/Alaskan Native, or Hispanic/Latino), substantially more diverse than US undergraduate \acCS majors, of which only 15.8% are from these underrepresented groups [14].

College students mentored K-12 students in their communities, including 300 high school students who compteed in a simplified High School Swarmathon in Netlogo. Additionally, the Swarmathon supported 60 students to attend workshops and the 2016 and 2017 Robotics Science and Systems (RSS) conferences and 17 students participated in research internships with mentors as other Swarmathon schools. All of these activities further developed pathways into STEM. As an example, 65% of RSS workshop participants indicated a desire to pursue a Ph.D. in robotics.

Ii Related Work

Ii-a \titlecapForaging Robot Swarms

Though swarm robot foraging has been studied analytically for decades [4], hardware implementations are rare [9]. For example, studies of \acCPF often only simulate the collection and transport of objects [15], and recent advances in hardware implementations, such as collaborative warehouse robots, require permanent infrastructure such as buried guide-wires or visual markers to operate [16].

Many attempts have been made to develop swarm algorithms and robot systems that address various aspects of the swarm foraging problem including scalability, energy efficiency, task allocation, and collection speed [17, 18, 19, 20, 21, 22]. However, autonomous swarm foraging remains a challenging open problem with no generally recognized best solution.

Projects using physical robot swarms include the Robotarium, a swarm robotics testbed providing remotely accessible robots and an arena [23]. This innovative project allows virtual experimentation with physical robot swarms, but it uses an overhead camera for localization and the robots themselves have no on-board sensors. Kilobots are simple, small robots intended to be integrated into large swarms. They have sensors and operate autonomously and collaboratively to push items through a maze or into specific configurations, but they have relatively limited mobility and operate in controlled laboratory environments [24]. The Swarmanoid project demonstrated successful implementation of a heterogeneous swarm whose robots collaborate in order to solve tasks like object retrieval in a highly specialized environment [25]. The challenges of operating physical swarms result in robots that are usually semi-autonomous in practice; they require frequent human support [26]. The reality gap between performance in simulation and performance in physical hardware [27] makes implementation in real robots particularly difficult. The hardware and software platforms for the Swarmies address these challenges so that \acCPF can be developed and subsequently tested in a system that is completely autonomous, operates without global knowledge or control, and can function outdoors.

Ii-B \titlecapEducational Robotics Environments and Competitions

Educational robotics is a growing field that is particularly effective at improving student performance in and attitudes towards STEM disciplines, and reducing gender and ethnic achievement gaps [28, 12, 29, 30, 31]. Examples for younger students include the distributed Robot Garden [32] and low-cost AERobots [33].

The \acNASA Robotic Mining Competition (RMC) is an annual engineering competition to design and build robots capable of mining the Martian surface as part of the \acISRU mission. Students focus on hardware design rather than autonomous control. Only 16% of the 900 RMC participants in 2017 identified as members of underrepresented groups222https://www.nasa.gov/offices/education/centers/kennedy/technology/nasarmc.html. The Swarmathon was designed to emulate the success and popularity of RMC while emphasizing robot autonomy and engagement of students from \acpMSI.

Ii-C \titlecapHardware

Teams selected for the Physical Competition are sent parts and instructions333https://github.com/BCLab-UNM/Swarmathon-Robot for building 3 Swarmies. Each robot has a front mounted gripper with actuated wrist and finger joints for grasping resources. Robots sense objects in the environment using three ultrasound range finders and a web camera. The ultrasounds have a range. The camera has a narrow field of view with a arc and range of .

Orientation and positional (pose) data are provided by a \acGPS unit and \acIMU with a magnetometer, along with wheel encoder odometry. Computation is provided by a small on-board computer (an Intel NUC). An Arduino Leonardo microcontroller provides the hardware interface between the on-board computer and both the wheel and gripper actuators and the encoders that measure wheel velocity. The robot body is built from laser cut and 3D printed components. The battery pack allows robots to run for 8 hours between charges, which with a default speed of speed of results in a range of . This allows a team of 6 Swarmies to search a linear distance of on a single charge, nearly the distance of a marathon that the remote-controlled Mars Rover Opportunity took 11 years to complete.

The resources that Swarmies locate and collect are cubes marked on all sides with AprilTags [34]. The perimeter of the central collection zone is also marked with AprilTags.

Ii-D \titlecapSoftware

Fig. 1: Swarmie Architecture. Software packages (shaded) interface with hardware through a USB connection, or with the Gazebo simulation.

The basic robot software, based on \acROS, is publicly available444https://github.com/BCLab-UNM/Swarmathon-ROS and provided to all teams. So teams may focus on the \acISRU challenge, we implement the low-level code that interfaces with physical and simulated actuators and sensors, performs necessary calibration and pre-configuration, and defines the simulated world for competition, including Swarmies, resources, and the search environment.

The software base is designed so that students can implement the same algorithms to control simulated or physical robots (Figure 1). The \acGUI and \acROS master either connect to the physical robots through a wireless network, or run in a Gazebo simulation. The diagnostic package monitors hardware components and alerts the user to problems. The robot pose is estimated using two \acpEKF that fuse encoder, \acIMU, and \acGPS data.

This base code includes: Gazebo simulation of the robots, target cubes, and competition arena (Figure 2); a \acGUI (Figure 1(b)) showing output of each hardware sensor, a map of the robots’ estimated positions (Figure 1(c)); customizable ROS packages to control behaviors (for search, communication, obstacle avoidance, and target pick up and drop off) and diagnostics (for sensors, actuators, wireless quality and the microcontroller); a \acROS package that interfaces with the Arduino to communicate with actuators and and sensors; open source packages including AprilTag and \acEKF packages.

The simple base code algorithm (modeled after the subsumption architecture [37]) is designed to be extended by teams to improve foraging. It implements simple collision avoidance, a random walk for search, and when a target appears in the camera view, the robot picks up the target and returns to the central collection zone to drop it off.

(a) Gazebo Simulation
(b) User Interface
(c) Example Search Pattern
Fig. 2: Gazebo and GUI. (a) Two robots bring target cubes marked with AprilTags to the collection zone. (b) Gazebo interface showing robot status, location tracing, and camera view. (c) Trace of the Montgomery College winning swarm of 6 robots; the dense activity shows robots traveling between a large target cluster and the collection zone.
Fig. 3: Physical Competition. Students compare strategies and cheer on Swarmies from the University of the District of Columbia as they drop cubes in the collection zone.

Iii Results

In the first 2 years, 29 of the 31 Physical teams and 20 of the 27 Virtual Teams successfully uploaded code to compete. The skills these students required for the competition included programming proficiency in \acROS, swarm algorithm design, and experimental design and testing.

Several teams developed effective new search algorithms that significantly improved the base code. The 2017 Physical Competition winner, \acSIPI, experimented with different search strategies inspired by snail shells, lawnmower paths, and spirograph-style geometric designs, ultimately deciding on a strategy inspired by the spokes of a bicycle wheel. \acSIPI’s algorithm displayed the greatest flexibility, performing at a high level across a variety of target distributions.

Figure 1(c) shows the Montgomery College search algorithm in an environment with one large cluster of target cubes. The zig-zag pattern shows the robots’ search path through the arena. Once the large cluster of resources was located, some robots continued searching while others focused on collecting from the cluster. This strategy successfully balanced the explore versus exploit trade-off and dramatically outperformed the other teams on nearly every distribution of resources to win the 2017 Virtual Competition. Engineering innovations (i.e., improving localization and resource pickup and drop-off) improved scores more than new search algorithms.

Iii-a \titlecapAssessed Student Learning Outcomes

263 students completed a comprehensive self-assessment (29%). 44 Faculty Mentors (76%) completed the survey with the same questions to evaluate their students’ performance.

Student growth was rated on 9 \acABET Criteria for Accrediting Computing Programs, and 2 \acABET Criteria for Accrediting Engineering Programs: “An ability to design and conduct experiments, as well as to analyze and interpret data,” and, “An ability to function on multidisciplinary teams.”555www.abet.org/accreditation/accreditation-criteria/ Students and faculty rated student degree of competency before and after the Swarmathon on a 4-point scale: none (1), low degree (2), moderate degree (3), and high degree (4).

Data for each year was analyzed with a dependent-samples t-test. Differences between the 2016 and 2017 cohorts were analyzed with an independent samples t-test. Table I shows the significant gains in self-assessed performance each year. 2017 gains were greater than 2016 gains in all categories and all results are statistically significant with .

There were no statistically significant differences in our assessment or survey data based on race, ethnicity, or gender, indicating that the Swarmathon provided a level playing field for competition, eliminating the achievement gap noted in other studies [31, 30]. Faculty mentor surveys closely mirrored student self-assessments on all but two ABET standards. Mentors ratings were more than 0.50 points lower than student ratings on (1) An understanding of professional, ethical, legal, security and social issues and responsibilities, and (2) An ability to analyze the local and global impact of computing on individuals, organizations, and society.

2016 2017
ABET Outcomes
Higher Degree Interest
Career Motivation
TABLE I: \titlecapChanges in student abilities and attitudes

Across both competition years, students also reported significant gains in their “desire or motivation to pursue a higher degree beyond the one I am currently pursuing,” and their “desire or motivation to pursue a career in \acCS, or Robotics, or other technical discipline in STEM” as seen in Table I. Gains were significant even among students who already possessed a moderate to high degree of interest in pursuing a more advanced degree.

The Swarmathon is effectively inspiring the next generation of engineers to tackle major scientific problems like the \acNASA \acISRU mission and to bring their diversity of background and experience into the \acSTEM workforce.

Acknowledgment

We thank Theresa Martinez, \acMUREP \acSTEM Engagement Manager; Paul Secor and Mary Baker of Secor Strategies, LLC; Kate Cunningham and Beatriz Palacios Abad of UNM ADVANCE; Elizabeth Esterly and the students of the Moses Biological Computation Lab, Vanessa Svihla, Aeron Haynie, the NASA event volunteers and UNM support staff, KSC Visitor Complex, and the Swarmathon students and Faculty Mentors.

\Urlmuskip

=0mu plus 1mu

References

  • [1] M. Brambilla, E. Ferrante, M. Birattari, and M. Dorigo, “Swarm robotics: a review from the swarm engineering perspective,” Swarm Intelligence, vol. 7, no. 1, pp. 1–41, 2013.
  • [2] O. Zedadra, N. Jouandeau, H. Seridi, and G. Fortino, “Multi-Agent Foraging: state-of-the-art and research challenges,” Complex Adaptive Systems Modeling, vol. 5, no. 1, p. 3, 2017.
  • [3] J. P. Hecker, K. Letendre, K. Stolleis, D. Washington, and M. E. Moses, “Formica ex machina: ant swarm foraging from physical to virtual and back again,” in Swarm Intelligence.   Springer, 2012, pp. 252–259.
  • [4] L. Steels, “Cooperation between distributed agents through self-organisation,” in Intelligent Robots and Systems’ 90.’Towards a New Frontier of Applications’, Proceedings. IROS’90. IEEE International Workshop on.   IEEE, 1990, pp. 8–14.
  • [5] A. Drogoul and J. Ferber, “From Tom Thumb to the Dockers: Some experiments with foraging robots,” From Animals to Animats, vol. 2, p. 451, 1993.
  • [6] A. Campo and M. Dorigo, “Efficient multi-foraging in swarm robotics,” in Advances in Artificial Life.   Springer, 2007, pp. 696–705.
  • [7] A. F. T. Winfield, “Towards an engineering science of robot foraging,” Distributed autonomous robotic systems, vol. 8, pp. 185–192, 2009.
  • [8] M. Dorigo, M. Birattari, and M. Brambilla, “Swarm robotics,” Scholarpedia, vol. 9, no. 1, p. 1463, 2014.
  • [9] A. F. T. Winfield, “Foraging Robots,” in Encyclopedia of Complexity and Systems Science.   Springer, 2009, vol. 6, ch. Foraging R, pp. 3682–3700.
  • [10] A. L. Griffith, “Persistence of women and minorities in STEM field majors: Is it the school that matters?” Economics of Education Review, vol. 29, no. 6, pp. 911–922, 2010.
  • [11] C. B. Leggon, “Women in science: Racial and ethnic differences and the differences they make,” The Journal of Technology Transfer, vol. 31, no. 3, pp. 325–333, 2006.
  • [12] S. D. Museus, R. T. Palmer, R. J. Davis, and D. Maramba, Racial and Ethnic Minority Student Success in STEM Education: ASHE Higher Education Report, Volume 36, Number 6.   John Wiley & Sons, 2011.
  • [13] US Department of Education, “FACT SHEET: Spurring African-American STEM Degree Completion [Press release],” 2016. [Online]. Available: https://www.ed.gov/news/press-releases/fact-sheet-spurring-african-american-stem-degree-completion
  • [14] S. Zweben and B. Bizot, “2016 Taulbee Survey: Generation CS Continues to Produce Record Undergrad Enrollment; Graduate Degree Production Rises at both Master’s and Doctoral Levels,” Computing Research News, vol. 29, no. 5, 2017.
  • [15] A. Brutschy, L. Garattoni, M. Brambilla, G. Francesca, G. Pini, M. Dorigo, and M. Birattari, “The TAM: abstracting complex tasks in swarm robotics research,” Swarm Intelligence, vol. 9, no. 1, pp. 1–22, 2015.
  • [16] J. Enright and P. R. Wurman, “Optimization and Coordinated Autonomy in Mobile Fulfillment Systems.” in Automated action planning for autonomous mobile robots, 2011, pp. 33–38.
  • [17] W. Liu, A. Winfield, J. Sa, J. Chen, and L. Dou, “Strategies for energy optimisation in a swarm of foraging robots,” in International Workshop on Swarm Robotics.   Springer, 2006, pp. 14–26.
  • [18] N. R. Hoff, A. Sagoff, R. J. Wood, and R. Nagpal, “Two foraging algorithms for robot swarms using only local communication,” in Robotics and Biomimetics (ROBIO), 2010 IEEE International Conference on.   IEEE, 2010, pp. 123–130.
  • [19] J. P. Hecker and M. E. Moses, “Beyond pheromones: evolving error-tolerant, flexible, and scalable ant-inspired robot swarms,” Swarm Intelligence, vol. 9, no. 1, pp. 43–70, 2015.
  • [20] E. Ferrante, A. E. Turgut, E. Duéñez-Guzmán, M. Dorigo, and T. Wenseleers, “Evolution of self-organized task specialization in robot swarms,” PLoS computational biology, vol. 11, no. 8, p. e1004273, 2015.
  • [21] Q. Lu, J. P. Hecker, and M. E. Moses, “The MPFA: A Multiple-Place Foraging Algorithm for Biologically-Inspired Robot Swarms,” IEEE/RSJ International Conference on Intelligent Robots and Systems, 2016.
  • [22] G. Fricke, J. Hecker, A. Griego, L. Tran, and M. Moses, “A distributed deterministic spiral search algorithm for swarms,” in IEEE International Conference on Intelligent Robots and Systems, 2016.
  • [23] D. Pickem, P. Glotfelter, L. Wang, M. Mote, A. Ames, E. Feron, and M. Egerstedt, “The Robotarium: A remotely accessible swarm robotics research testbed,” in Robotics and Automation (ICRA), 2017 IEEE International Conference on.   IEEE, 2017, pp. 1699–1706.
  • [24] M. Rubenstein, C. Ahler, N. Hoff, A. Cabrera, and R. Nagpal, “Kilobot: A low cost robot with scalable operations designed for collective behaviors,” Robotics and Autonomous Systems, vol. 62, no. 7, pp. 966–975, 2014.
  • [25] M. Dorigo, D. Floreano, L. M. Gambardella, F. Mondada, S. Nolfi, T. Baaboura, M. Birattari, M. Bonani, M. Brambilla, and A. Brutschy, “Swarmanoid: a novel concept for the study of heterogeneous robotic swarms,” IEEE Robotics & Automation Magazine, vol. 20, no. 4, pp. 60–71, 2013.
  • [26] A. Rosenfeld, N. Agmon, O. Maksimov, and S. Kraus, “Intelligent agent supporting human-multi-robot team collaboration,” Artificial Intelligence, 2017.
  • [27] N. Jakobi, P. Husbands, and I. Harvey, “Noise and the reality gap: The use of simulation in evolutionary robotics,” in European Conference on Artificial Life.   Springer, 1995, pp. 704–720.
  • [28] S. Freeman, S. L. Eddy, M. McDonough, M. K. Smith, N. Okoroafor, H. Jordt, and M. P. Wenderoth, “Active learning increases student performance in science, engineering, and mathematics,” Proceedings of the National Academy of Sciences, vol. 111, no. 23, pp. 8410–8415, 2014.
  • [29] A. G. Welch, “Using the TOSRA to Assess High School Students’ Attitudes toward Science after Competing In the FIRST Robotics Competition: An Exploratory Study.” Eurasia Journal of Mathematics, Science & Technology Education, vol. 6, no. 3, 2010.
  • [30] A. S. Richmond, G. A. Boysen, and R. A. R. Gurung, An Evidence-based Guide to College and University Teaching: Developing the Model Teacher.   Routledge, 2016.
  • [31] C. M. Steele, Whistling Vivaldi: How Stereotypes Affect Us and What We Can Do.   New York, NY, US: WW Norton & Co, 2010.
  • [32] L. Sanneman, D. Ajilo, J. DelPreto, A. Mehta, S. Miyashita, N. A. Poorheravi, C. Ramirez, S. Yim, S. Kim, and D. Rus, “A Distributed Robot Garden System,” in Robotics and Automation (ICRA), 2015 IEEE International Conference on.   IEEE, 2015, pp. 6120–6127.
  • [33] M. Rubenstein, B. Cimino, R. Nagpal, and J. Werfel, “AERobot: An affordable one-robot-per-student system for early robotics education,” in Robotics and Automation (ICRA), 2015 IEEE International Conference on.   IEEE, 2015, pp. 6107–6113.
  • [34] E. Olson, “AprilTag: A robust and flexible visual fiducial system,” in Robotics and Automation (ICRA), 2011 IEEE International Conference on.   IEEE, 2011, pp. 3400–3407.
  • [35] M. Quigley, K. Conley, B. Gerkey, J. Faust, T. Foote, J. Leibs, R. Wheeler, and A. Y. Ng, “ROS: an open-source Robot Operating System,” in ICRA workshop on open source software, vol. 3, no. 3.2.   Kobe, 2009, p. 5.
  • [36] T. Moore and D. Stouch, “A Generalized Extended Kalman Filter Implementation for the Robot Operating System,” in Proceedings of the 13th International Conference on Intelligent Autonomous Systems (IAS-13).   Springer, 2014, pp. 335–348.
  • [37] R. Brooks, “A robust layered control system for a mobile robot,” IEEE journal on robotics and automation, vol. 2, no. 1, pp. 14–23, 1986.
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
""
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
   
Add comment
Cancel
Loading ...
198682
This is a comment super asjknd jkasnjk adsnkj
Upvote
Downvote
""
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters
Submit
Cancel

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test
Test description