Robot Ecology: An Inspiration for Future Ecologists

Americans value science, have a “great deal of confidence” in the scientific community, and believe that scientific pursuits create opportunities for the next generation (NSF 2020). It is surprising, then, that the public has not heeded calls to preserve our natural world and planetary functioning. Scientists have shown that the human alteration of ecosystems is global, that few realms of life are unaffected (Halpern et al. 2015), and that these effects will not only affect Earth functioning but are an existential threat to humans (Steffen et al. 2015). However, increased insulation from ecological realities and natural processes have culminated in people perceiving that they are apart from the natural world and in inaction on some of the world's most pressing environmental challenges (Nadkarni and Stasch 2013). This growing dissonance between people and nature has forced ecologists to engage the public in new and exciting ways and approaches that avoid confrontation (Nadkarni 2013) and pessimism (McAfee et al. 2019) and that instead offers solutions, hopes, and even inspiration.

Engineers and roboticists have recently embraced ecological inspiration to advance novel lines of technological inquiry (Egerstedt et al. 2018). One outcome of this new perspective in robotics is that the traditional design approach, with a singular goal or task in mind, has been revisited. Robots have typically been assessed by performance with a reward function to be maximized, such as the accuracy and repeatability of industrial robots in manufacturing or the speed at which fetch-and-carry robots deliver an item in a warehouse or the precision of surgical robot following a predefined trajectory. For robots outside of a factory, warehouse, or controlled laboratory—those deployed over long time periods and to natural environments for monitoring, precision agriculture, or long-distance exploration—“surviving” in the environment takes precedence even over maximizing any particular reward function (Egerstedt et al. 2018). This shift in design paradigm entails that certain “survival sets” are rendered forward invariant, meaning that if the robot starts out in the set, it remains in that set for all times. Such considerations are encoded in terms of constraints rather than goals, which is where robotics and ecology conjugate: environmental constraints sculpting and molding adaptive traits for individual success (Egerstedt et al. 2018).

The SlothBot is the first manifestation of the survival concept in robotics and a new conceptualization of robot ecology (Egerstedt et al. 2018). It resulted from a unique collaboration of roboticists and ecologists, and is explicitly designed to take advantage of ecological knowledge of species persisting in an energetically constrained environment (figure 1). Tree sloths have evolved an exquisite suite of adaptive traits to minimize energetic expenditure: They possess the lowest metabolic rate for a mammal (Pauli et al. 2016), anatomical specializations to minimize transportation costs (Cliffe et al. 2014) and efficiently digest food (Dill-McFarland et al. 2016), reduced activity patterns, a spatially anchored and small home range, and unique behaviors to acquire resources and thermoregulate (Pauli et al. 2014). Together, these traits enable sloths to survive a threatening and energetically limited landscape. The SlothBot mirrors sloths in being suspended on wires in the forest canopy, being slow and energy efficient. It is solar powered and maintains sufficient separation from obstacles in the environment so that the robot always has enough battery charge to reach a sunny spot to recharge and, spatially anchored, return to locations where measured data can be downloaded. As a result of adhering to this constraint-based design, the SlothBot first survives autonomously and then is designed to measure environmental variables.

Figure 1.

Three-toed sloth resting on a cable in an agroecosystem in Costa Rica, which was the inspiration for roboticists to develop the Slothbot, shown traversing an installed cable in the Atlanta Botanical Gardens.

Since the deployment of the SlothBot to the Atlanta Botanical gardens, there has been a surge of public interest in these two charismatic and inspiring animated entities: autonomous robots and living tree sloths. Indeed, SlothBots were picked up by dozens of international and diverse news outlets (e.g., BBC, CNN, The Independent, Nature World News, Science Daily, The Costa Rica News, Electronics Weekly, and even youth outlets such as The Week Junior) reaching a potential audience of nearly 500 million viewers. In August 2020, when CNN featured the SlothBot their homepage, CNN reported 176 million unique visitors to their site. Such interest should not be surprising: Sloths are charismatic; there's an international sloth day, sloths are featured on Costa Rican currency, and they are used in tourism marketing in Latin America. Robots too are inspirational, imagined droids in sci-fi movies and working robots such as NASA Mars rovers, self-driving cars, and robotic pets and household cleaners. Integrating ecological insights with human-made designs in engineering and robotics not only produces interesting and novel artifacts but can promote technological advancements in robotics. Perhaps most importantly, such cross-disciplinary endeavors can broaden ecological awareness and act as an accessible portal for new generations of scientists and different segments of the public to be inspired by our natural world.