Letter | Published:

Designing minimal and scalable insect-inspired multi-locomotion millirobots

Abstract

In ant colonies, collectivity enables division of labour and resources1,2,3 with great scalability. Beyond their intricate social behaviours, individuals of the genus Odontomachus4, also known as trap-jaw ants, have developed remarkable multi-locomotion mechanisms to ‘escape-jump’ upwards when threatened, using the sudden snapping of their mandibles5, and to negotiate obstacles by leaping forwards using their legs6. Emulating such diverse insect biomechanics and studying collective behaviours in a variety of environments may lead to the development of multi-locomotion robotic collectives deployable in situations such as emergency relief, exploration and monitoring7; however, reproducing these abilities in small-scale robotic systems with simple design and scalability remains a key challenge. Existing robotic collectives8,9,10,11,12 are confined to two-dimensional surfaces owing to limited locomotion, and individual multi-locomotion robots13,14,15,16,17 are difficult to scale up to large groups owing to the increased complexity, size and cost of hardware designs, which hinder mass production. Here we demonstrate an autonomous multi-locomotion insect-scale robot (millirobot) inspired by trap-jaw ants that addresses the design and scalability challenges of small-scale terrestrial robots. The robot’s compact locomotion mechanism is constructed with minimal components and assembly steps, has tunable power requirements, and realizes five distinct gaits: vertical jumping for height, horizontal jumping for distance, somersault jumping to clear obstacles, walking on textured terrain and crawling on flat surfaces. The untethered, battery-powered millirobot can selectively switch gaits to traverse diverse terrain types, and groups of millirobots can operate collectively to manipulate objects and overcome obstacles. We constructed the ten-gram palm-sized prototype—the smallest and lightest self-contained multi-locomotion robot reported so far—by folding a quasi-two-dimensional metamaterial18 sandwich formed of easily integrated mechanical, material and electronic layers, which will enable assembly-free mass-manufacturing of robots with high task efficiency, flexibility and disposability.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Data availability

All data generated or analysed during this study are included in the published article, and are available from the corresponding author on reasonable request.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Heyman, Y., Shental, N., Brandis, A., Hefetz, A. & Feinerman, O. Ants regulate colony spatial organization using multiple chemical road-signs. Nat. Commun. 8, 15414 (2017).

  2. 2.

    Gordon, D. M. The ecology of collective behavior. PLoS Biol. 12, e1001805 (2014).

  3. 3.

    Franks, N. R. & Richardson, T. Teaching in tandem-running ants. Nature 439, 153 (2006).

  4. 4.

    Sorger, D. M. & Zettel, H. On the ants (Hymenoptera: Formicidae) of the Philippine Islands: V. The genus Odontomachus Latreille, 1804. Myrmecol. News 14, 141–163 (2011).

  5. 5.

    Patek, S. N., Baio, J. E., Fisher, B. L. & Suarez, A. V. Multifunctionality and mechanical origins: ballistic jaw propulsion in trap-jaw ants. Proc. Natl Acad. Sci. USA 103, 12787–12792 (2006).

  6. 6.

    Sorger, D. M. Snap! Trap-jaw ants in Borneo also jump using their legs. Front. Ecol. Environ. 13, 574–575 (2015).

  7. 7.

    Fearing, R. S. Challenges for effective millirobots. In IEEE Int. Sym. Micro-NanoMech. Hum. Sci. 1–5 (IEEE, 2006).

  8. 8.

    Rubenstein, M., Cornejo, A. & Nagpal, R. Programmable self-assembly in a thousand-robot swarm. Science 345, 795–799 (2014).

  9. 9.

    Kernbach, S. et al. Adaptive collective decision-making in limited robot swarms without communication. Int. J. Robot. Res. 32, 35–55 (2013).

  10. 10.

    Arvin, F., Murray, J., Zhang, C. & Yue, S. Colias: an autonomous micro robot for swarm robotic applications. Int. J. Adv. Robot. Syst. 11, 113 (2014).

  11. 11.

    Pickem, D. et al. The Robotarium: a remotely accessible swarm robotics research testbed. In 2017 IEEE Int. Conf. Rob. Autom. 1699–1706 (IEEE, 2016).

  12. 12.

    Weston-Dawkes, W. P., Ong, A. C., Majit, M. R. A., Joseph, F. & Tolley, M. T. Towards rapid mechanical customization of cm-scale self-folding agents. In IEEE Int. Conf. Intel. Rob. Sys. 4312–4318 (IEEE, 2017).

  13. 13.

    Jung, G.P., Casarez, C. S., Jung, S. P., Fearing, R. S. & Cho, K. J. An integrated jumping–crawling robot using height-adjustable jumping module. In 2016 IEEE Int. Conf. Rob. Autom. 4680–4685 (IEEE, 2016).

  14. 14.

    Zhao, J., Yan, W., Xi, N., Mutka, M. W. & Xiao, L. A miniature 25 grams running and jumping robot. In 2014 IEEE Int. Conf. Rob. Autom. 5115–5120 (IEEE, 2014).

  15. 15.

    Morrey, J. M., Lambrecht, B., Horchler, A. D., Ritzmann, R. E. & Quinn, R. D. Highly mobile and robust small quadruped robots. In 2003 IEEE/RSJ Int. Conf. Intel. Rob. Sys. 1, 82–87 (IEEE, 2003).

  16. 16.

    Song, G., Yin, K., Zhou, Y. & Cheng, X. A surveillance robot with hopping capabilities for home security. IEEE Trans. Consum. Electron. 55, 2034–2039 (2009).

  17. 17.

    Zhang, Y., Zhang, L., Wang, W., Li, Y. & Zhang, Q. Design and implementation of a two-wheel and hopping robot with a linkage mechanism. IEEE Access 6, 42422–42430 (2018).

  18. 18.

    Bertoldi, K., Vitelli, V., Christensen, J. & van Hecke, M. Flexible mechanical metamaterials. Nat. Rev. Mater. 2, 17066 (2017).

  19. 19.

    Larabee, F. J. & Suarez, A. V. Mandible-powered escape jumps in trap-jaw ants increase survival rates during predator–prey encounters. PLoS One 10, e0124871 (2015).

  20. 20.

    Bergbreiter, S. Effective and efficient locomotion for millimeter-sized microrobots. In 2008 IEEE/RSJ Int. Conf. Intel. Rob. Sys. 4030–4035 (IEEE, 2008).

  21. 21.

    Haldane, D. W., Plecnik, M. M., Yim, J. K. & Fearing, R. S. Robotic vertical jumping agility via series-elastic power modulation. Sci. Rob. 1, eaag2048 (2016).

  22. 22.

    Hu, W., Lum, G. Z., Mastrangeli, M. & Sitti, M. Small-scale soft-bodied robot with multimodal locomotion. Nature 554, 81–85 (2018).

  23. 23.

    Kim, Y., Yuk, H., Zhao, R., Chester, S. A. & Zhao, X. Printing ferromagnetic domains for untethered fast-transforming soft materials. Nature 558, 274–279 (2018).

  24. 24.

    Jäger, P. Cebrennus Simon, 1880 (Araneae: Sparassidae): a revisionary up-date with the description of four new species and an updated identification key for all species. Zootaxa 3790, 319–356 (2014).

  25. 25.

    Zhakypov, Z., Huang, J. L. & Paik, J. A novel torsional shape-memory alloy actuator: modeling, characterization, and control. IEEE Robot. Autom. Mag. 23, 65–74 (2016).

  26. 26.

    Wood, R.J, Avadhanula, S., Sahai, R., Steltz, E. & Fearing R. S. Microrobot design using fiber-reinforced composites. J. Mech. Des. 130, 052304 (2008).

  27. 27.

    Zhakypov, Z. & Paik J. Design methodology for constructing multi-material origami robots and machines. IEEE Trans. Rob. 34, 151–165 (2018).

  28. 28.

    Felton, S., Tolley, M. Demaine, E., Rus, D. & Wood, R. J. A method for building self-folding machines. Science 345, 644–646 (2014).

  29. 29.

    Bennet-Clark, H. C. The energetics of the jump of the locust Schistocerca gregaria. J. Exp. Biol. 63, 53–83 (1975).

  30. 30.

    Nabawy, M. R., Sivalingam, G., Garwood, R. J., Crowther, W. J. & Sellers, W. I. Energy and time optimal trajectories in exploratory jumps of the spider Phidippus regius. Sci. Rep. 8, 7142 (2018).

Download references

Acknowledgements

We thank H. Aonuma from the Complex Systems Research Group of Hokkaido University for providing insight into the behaviours of the trap-jaw ant. K.M. acknowledges financial support from the Japan Public–Private Partnership Student Study Abroad Program. This work is supported by the Swiss National Science Foundation (SNSF) ‘START’ Project and the Swiss National Center of Competence in Research (NCCR) Robotics.

Reviewer information

Nature thanks Adam Stokes and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Z.Z., K.M., K.H. and J.P. designed the study and interpreted the results. Z.Z. and J.P. conceived the idea of developing multi-locomotion millirobot collectives. Z.Z. and K.M. developed the millirobot hardware platform. Z.Z. designed the multi-locomotion mechanisms and models, and K.M. designed the electronics and communication. Z.Z. produced the figures and videos. Z.Z. and J.P. wrote the manuscript with input from all authors.

Competing interests

The authors declare no competing interests.

Correspondence to Jamie Paik.

Extended data figures and tables

Extended Data Fig. 1 Design challenges and needs of biological versus artificial multi-locomotion collectives.

Trap-jaw ant collectivity, backed by scalable reproduction and the minimal multi-locomotion mechanisms that are integrated into their jaws and legs are the key to their survival in cluttered environments, which have emerged from evolutionary processes. Replicating these abilities in engineered systems will enable the use of millirobots in applications such as emergency mitigation, environmental monitoring and exploration with high task flexibility and efficiency. However, constructing minimal, integrated multi-locomotion mechanisms remains a major challenge for robotic hardware design that, when addressed, will enable robot miniaturization and assembly-free mass production for collective implementations. CAM, computer-aided manufacturing.

Extended Data Fig. 2 Free-body diagrams of Tribot for calculating the locomotion kinematics and dynamics for all five locomotion gaits.

a, Tribot transitioning from initial stance to take-off, applicable for the distance- and somersault-jump gaits and walking gaits as a result of snap-through at the Y-hinge side. The Y-hinge is modelled as a revolute pin joint connecting three links. The snap-through motion generated by compression of the SMA spring actuator (k3) instantly rotates the side legs (1 and 2) and pushes the rear leg (3) against the ground. This produces a ground reaction force that lifts the robot in a ballistic projectile motion with take-off velocity v0. For somersault jumping, the bottom spring actuator (k1) activates shortly before the side spring (k3) to achieve free-body rotation during flight. b, The robot can perform height jumps on any three edges; however, for reaching high altitudes, it is most effective on the edge without rubber friction pads (which are located on legs 2 and 3). Here, the snap-through occurs at the Y-hinge bottom, and the rapid closing of the bottom legs produces a ground reaction force that lifts the robot up vertically. c, The crawling locomotion occurs on the edge with latches; the robot moves by opening and closing the bottom legs (2 and 3) using stick-slip motion. GRF, ground reaction force.

Extended Data Fig. 3 The proximity measurement data for the division-of-labour experiment and the event chart for the tandem-running experiment.

a, The proximity data measured by the monitor robot, showing the linear propagation of the workers with each pushing step. The object is moved its set distance, 30 mm. b, The event chart for the leader–follower tandem-running experiment with obstacle avoidance by communication.

Extended Data Table 1 Tribot’s functional components and mass budget
Extended Data Table 2 Tribot’s locomotion performance under different conditions
Extended Data Table 3 Comparison of Tribot with small-scale terrestrial multi-locomotion robots and insects

Supplementary information

Video 1

The demonstration of Tribot height, distance and somersault jumping gaits The first part shows the height jumping capacity of the robot on two different edges: with and without latches (Experiments 1 and 2). In the second part, the robot executes the distance jump with high and low trigger power, and a 5 g payload (Experiments 3, 4 and 5). The video also demonstrates Tribot’s ability to jump over an obstacle and onto a stair. The last part shows Tribot’s somersault jumping capability by flipping in the air after jumping (Experiment 6). In each experiment, the robot is programmed for set, stance and trigger configurations with different sequence of SMA activation.

Video 2

The demonstration of flic-flac walking locomotion of Tribot on different terrain. The robot walks on flat surface and rough terrain composing of raisin-sized stones (Experiments 7 and 8). Tribot also walks onto a slope of 10° (Experiment 9). In all cases, the robot is placed in a transparent channel, slightly wider than its width, to constrain its deviation in lateral direction. It is controlled remotely to perform three consecutive rolling steps.

Video 3

The demonstration of Tribot inchworm crawling locomotion on different terrain. The robot effectively crawls on flat surface and onto a 10º ramp (Experiments 10 and 11).

Video 4

The demonstration of Tribot multi-locomotion ability in parkour scenario. Tribot first crawls over the flat section and then walks on the rough section until it reaches the obstacle made of large stones and jumps over. (Experiments 10 and 11).

Video 5

The demonstration of Tribot collective operation in division of labor and tandem running scenarios. For the first scenario, five robots are allocated leader, worker, monitor and messenger roles. Here, two workers push an object to a target position in the direction of the monitor, which measures the relative position of the object with respect to itself using a proximity sensor. The messenger ensures two-way communication link between the leader and the monitor due to signal interruption by the object. This highlights the benefits of robot scalability in manipulation and communication with limited resources. For the second task, two robots are given leader and follower roles to overcome a gap obstacle in tandem. The leader is programmed to continuously scan for obstacles, while the follower follows the leader blindly with no energy expense for scanning. The leader gets into the gap, detects it and informs the follower about the upcoming obstacle, and jumps over. The follower calculates number of steps to crawl and then jumps to avoid the gap. It highlights the benefits of multi-locomotion ability for robotic collectives in overcoming obstacles.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark
Fig. 1: Design and fabrication of the trap-jaw-ant-inspired Tribot multi-locomotion millirobot.
Fig. 2: Individual multi-locomotion experiments and their combination in the parkour scenario.
Fig. 3: Locomotion performance of Tribot in different conditions and its COT compared to robots and insects.
Fig. 4: Collective labour experiments.
Extended Data Fig. 1: Design challenges and needs of biological versus artificial multi-locomotion collectives.
Extended Data Fig. 2: Free-body diagrams of Tribot for calculating the locomotion kinematics and dynamics for all five locomotion gaits.
Extended Data Fig. 3: The proximity measurement data for the division-of-labour experiment and the event chart for the tandem-running experiment.

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.