Robotics Simulator for Development and Verification of Swarm Behaviors

摘要

Abstract Swarm control strategies allow for decentralized control of many simple robots to perform collective behaviors based on local interactions. We have created a new platform for developing and exploring swarm behaviors, supporting both simulation and experimental verification. The platform, designed entirely in Simulink, provides a simple implementation with satisfactory dynamics to replicate experimental trials on Santa Clara University’s Robotic Systems Laboratory’s Low-Cost Indoor Testbed. The platform allows for seamless transitions from simulation to experimentation on the testbed. The platform is currently equipped with attract, disperse, obstacle avoidance, and go-to behaviors which have been verified both in simulation and on the experimental testbed. Additionally, a composite behavior, flocking, composed of attract and go-to behaviors is presented to demonstrate the flexibility of the simulator. Future work will expand the available behaviors to include swarm adaptive navigation behavior primitives like minima and maxima finding as well as contour and ridge following.