Growing cities are a world-wide phenomenon and simultaneously awareness about potential dangers due to air pollution, heat, and pathogens is increasing. Integrated and permanent monitoring of environmental features in cities can help to establish an early warning system and to provide data for policy makers. In our new project ‘WatchPlant,’ we propose a green approach for urban monitoring by a network of sensors tightly coupled with natural plants. We want to develop a sustainable, energy-efficient bio-hybrid system that harvests energy from living plants and utilizes methods of phytosensing, that is, using natural plants as sensors. We present our concept, here with focus on Alife-related methods operating on the gathered plant data and the bio-hybrid network. With a self-organizing network of sensors, that are alive, we hope to contribute to our future of livable green cities.

Robot-plant bio-hybrid systems are getting increasing attention due to the wide range of applications they offer. Such synergies between robots and natural plants will allow, for example, establishing highly reliable environmental monitoring systems or growing the architecture of our future cities. We explore the latter application where robots exploit the plants’ ability to produce construction material, and plants exploit the robots’ sensing and computational capabilities. In our previous work, we used machine learning techniques to model plant behavior in their early life stages. We collected a 10-point plant stem description dataset and used it to train an LSTM as a forward model that predicts plant dynamics and drives the evolution of plant shaping controllers. Here, we show our vision to model plant behaviors in later stages, where full-plant morphology will be used to train state-of-the-art sequence modeling networks capable of simulating more complex plant dynamics.

Robot swarms are known to be robust to individual robot failures. However, a reduced swarm size causes a reduced swarm density. A too low swarm density may then decrease swarm performance, that should be compensated by adapting the individual behavior. Similarly, swarm behaviors can also be adapted to changes in the environment, such as dynamic light conditions. We study aggregation of swarm robots controlled by an extended variant of the BEECLUST algorithm. The robots are asked to aggregate at the brightest spot in their environment. Our approach efficiently adapts this swarm aggregation behavior to variability in swarm density and light conditions. First, each robot individually monitors its environment continuously by sampling its local swarm density and perceived light condition. Second, we exploit the collaboration of robots by letting them share features of these measurements with their neighbors by communication. In extensive robot swarm experiments with ten robots we validate our approach with dynamically changing swarm densities and under dynamic light conditions. We find an improved performance compared to robot swarms without communication and without awareness of the swarm density.