FocusCOLLECTIVE BEHAVIOR

Reflections on the future of swarm robotics

See allHide authors and affiliations

Science Robotics  09 Dec 2020:
Vol. 5, Issue 49, eabe4385
DOI: 10.1126/scirobotics.abe4385

Tables

  • Table 1 Timeline of swarm robotics research.

    Milestones reached by swarm robotics research in the past and projections of future developments. For each milestone, we indicate whether it entails innovation in software (SW), hardware (HW), or both.

    1990–2000A new paradigm is tested in which collaboration is emergent from simple (often bioinspired)
    behaviors. First experiments with robots demonstrating self-organization by means of indirect
    (stigmergic) and local interactions, with clear inspiration from swarm intelligence.
    SW
    2000–2005The possibility to design robots cooperating in a swarm is extended to several new tasks, entailing
    manipulation of objects, task allocation, and tasks that strictly require collaboration in order to be
    solved.
    SW
    2002–2006The Swarm-bots project demonstrates robot swarms capable of self-assembly. Robots are capable of
    building pulling chains and large structures that can transport heavy weights and deal with
    rough terrain.
    HW and SW
    2004–2008Initial demonstrations of the automatic design of robot swarms by means of evolutionary algorithms,
    leading to the establishment of the evolutionary swarm robotics approach.
    SW
    2005–2009First attempts at developing standard swarm robotics platforms (e-pucks) and miniature robots for
    swarm robotics research (Alice, Jasmine).
    HW
    2006–2010The Swarmanoid project demonstrates for the first time heterogeneous robot swarms composed of
    three groups of robots: flying, climbing, and ground-based robots.
    HW and SW
    2010–2015Different approaches appear for the design of robot swarms: Advanced methods for automatic design
    (AutoMoDe, novelty search), design patterns, mean-field models, and optimal stochastic approaches.
    SW
    2014–2019The “control without computation” approach develops swarm robotics behaviors with direct
    sensor-actuator mapping and no computation whatsoever.
    SW
    2016–2020Swarms of flying drones become available for research, and decentralized solutions are studied and
    deployed.
    HW and SW
    2020–2025First demonstration of robot swarms capable of autonomously learning a suitable collective behavior for a
    given class of problems.
    SW
    2020–2030First civil applications of robot swarms to precision agriculture and infrastructure inspection and
    maintenance. Military applications largely use non-combat unmanned drones to cooperatively accomplish
    information gathering and mission support actions.
    HW and SW
    2025–2030Deployment of robot swarms for maritime and deep-sea applications, providing support to ecological
    monitoring, surveillance, and fishing.
    HW
    2025–2035The entertainment sector uses robot swarms for interactive, immersive displays. Robot swarms are
    employed within the city, sharing the environment with operators and citizens. Robots will be insect- or
    pet-like devices that will collaborate to carry out service tasks such as cleaning, grazing, or delivering goods.
    HW and SW
    2030–2040First space exploration mission on the Moon and Mars with miniature rover swarms, expanding the
    explored area and demonstrating on-site construction abilities.
    HW
    2030–2045Millimeter-scale soft-bodied robot swarms enter agricultural fields for pest control or aquatic environments
    to collect microplastics.
    HW and SW
    2035–2050Microscopic robot swarms are demonstrated for medical applications such as targeted drug delivery, and
    clinical trials with human participants begin.
    HW and SW

Stay Connected to Science Robotics

Navigate This Article