Abstract
The design of robust yet simple communication mechanisms, that allow the cooperation through direct interaction among robots, is an important aspect of swarm robotics systems. In this paper, we analyze how an identical continuous-time recurrent neural network (CTRNN) controller can lead to the emergence of different kinds of communications within the swarm, either abstract or situated, depending on the problem to be faced. More precisely, we address two swarm robotics tasks that require, at some extent, communication to be solved: leader selection and borderline identification. The parameters of the CTRNN are evolved using separable natural evolution strategies. It is shown that, using the same starting conditions and robots’ controllers, the evolution process leads to the emergence of utterly diverging communications. Firstly, an abstract communication, in which the message carries all the information, results from evolution in the leader selection task. Alternatively, a purely situated communication, meaning that only the context is communicative, emerges when dealing with the borderline identification problem. Nonetheless, scalability and robustness properties are successfully validated.
Highlights
Swarm robotics (SR) [1] is the research field that, combining aspects of artificial intelligence and robotics, studies the use of many simple distributed robots that collectively cooperate in order to solve complex tasks
- Scalability: In each experiment, the behavior is assessed for different swarm sizes. - Robustness: Robustness is evaluated by means of introducing an alteration in the task at some point in time during the simulation. - Communication: The emerged communication is described for each problem
The borderline identification is a task in which swarm members have to detect if they are in the borderline or shape of the swarm or not
Summary
Swarm robotics (SR) [1] is the research field that, combining aspects of artificial intelligence and robotics, studies the use of many simple distributed robots that collectively cooperate in order to solve complex tasks. In this paper we explore the use of Separable Natural Evolution Strategies (SNES, see [18]), in order to evolve CTRNN neural controllers for different tasks Another critical design step in swarm robotics is the communication mechanics of the group. We consider that the proposed tasks are remarkably suitable benchmark experiments, in order to validate the previously mention hypothesis of communication emergence, due to the following reasons: (i) their successful completion requires communication, (ii) the required semantics of both problems are expected to be different and (iii) both tasks can be addressed using the same artificial neural network as robot’s controller (because the robot sensors and actuators are common to both experiments). The source code developed and used for this paper is available at https://github.com/r-sendra/SpikeSwarmSim accessed on 22 Janunary 2021
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
