Behavior Trees as a Control Architecture in the Automatic Modular Design of Robot Swarms

Abstract

Previous research has shown that automatically combining low-level behaviors into a probabilistic finite state machine produces control software that crosses the reality gap satisfactorily. In this paper, we explore the possibility of adopting behavior trees as an architecture for the control software of robot swarms. We introduce Maple: an automatic design method that combines preexisting modules into behavior trees. To highlight the potential of this control architecture, we present robot experiments in which we compare Maple with Chocolate and EvoStick on two missions: foraging and aggregation. Chocolate and EvoStick are two previously published automatic design methods. Chocolate is a modular method that generates probabilistic finite state machines and EvoStick is a traditional evolutionary robotics method. The results of the experiments indicate that behavior trees are a viable and promising architecture to automatically generate control software for robot swarms.JK and AL contributed equally to the research and should be considered co-first authors. Behavior trees were originally brought to the attention of the authors by DB. The proposed method was conceived by the four authors. It was implemented and tested by JK and AL. The initial draft of the manuscript was written by JK and AL and then revised by DB and MB. The research was directed by MB.The final authenticated version is available online at https://doi.

Simulation-only experiments to mimic the effects of the reality gap in the automatic design of robot swarms

“…The two methods-AutoMoDe-Vanilla and AutoMoDe-Chocolate-have shown to produce control software that crosses the reality gap more satisfactorily than those produced by EvoStick, an implementation of the traditional neuro-evolutionary robotics (Francesca et al 2014(Francesca et al , 2015. These results were further confirmed by follow up studies (Kuckling et al 2018;Hasselmann et al 2018b).…”

Concurrent design of control software and configuration of hardware for robot swarms under economic constraints

“…In AutoMoDe, control software is generated by automatically assembling predefined modules and by fine-tuning their free parameters. A number of methods have been proposed that belong to AutoMoDe: Vanilla (Francesca et al, 2014), Chocolate (Francesca et al, 2015), Gianduja (Hasselmann, Robert & Birattari, 2018), and Maple (Kuckling et al, 2018). Each of these methods is characterized by a specific set of predefined modules, a software architecture into which these modules can be combined, and an optimization algorithm that searches the space of the possible ways in which modules can be combined into the given architecture and the space of the free parameters.…”

“…In this research, we introduce Waffle, a new instance of the AutoMoDe family of automatic design methods. All previously published instances of AutoMoDe generate control software for the e-puck platform (Mondada et al, 2009) by selecting, combining, and fine-tuning predefined, mission-independent software modules (Francesca et al, 2014;Francesca et al, 2015;Kuckling et al, 2018;Hasselmann, Robert & Birattari, 2018). Waffle is based on Chocolate (Francesca et al, 2015).…”

Peer Review #1 of "Concurrent design of control software and configuration of hardware for robot swarms under economic constraints (v0.1)"

“…Chocolate (Francesca et al, 2015), Gianduja (Hasselmann et al, 2018b), and Maple (Kuckling et al, 2018). Each of these methods is characterized by a specific set of predefined modules, a software architecture into which these modules can be combined, and an optimization algorithm that searches the space of the possible ways in which modules can be combined into the given architecture and the space of the free parameters.…”

“…In this research, we introduce Waffle, a new instance of the AutoMoDe family of automatic design methods. All previously published instances of AutoMoDe generate control software for the e-puck platform (Mondada et al, 2009) by selecting, combining, and fine-tuning predefined, mission-independent software modules (Francesca et al, 2014(Francesca et al, , 2015Kuckling et al, 2018;Hasselmann et al, 2018b). Waffle is based on Chocolate (Francesca et al, 2015).…”