Design of a Software-in-the-Loop Simulation System for Micro Aerial Vehicle Swarming
Gao, Changyu (2017-08-15)
Design of a Software-in-the-Loop Simulation System for Micro Aerial Vehicle Swarming
Gao, Changyu
(15.08.2017)
Tätä artikkelia/julkaisua ei ole tallennettu UTUPubiin. Julkaisun tiedoissa voi kuitenkin olla linkki toisaalle tallennettuun artikkeliin / julkaisuun.
Turun yliopisto
Kuvaus
Siirretty Doriasta
Tiivistelmä
Over the past decade, micro aerial vehicles (MAV) have experienced a vigorous success in both research and business. With the continuous miniaturization of MAV, many research institutes are focusing on the study of MAV swarm control algorithms, communication topology, task allocation and other new issues. Moreover, the MAV swarm has a great military value for the army.
Compared with the research of a single MAV, the study of an MAV swarm is more complicated and dangerous. Because multiple MAVs are likely to collide and interfere with each other, their swarm behavior is harder to predict. This paper presents a Software-in-the-Loop (SIL) simulation system for MAV swarming research. This system is able to test the MAV swarm in a software simulated environment.
he SIL simulation system is based on the robot operating system (ROS) and the three-dimensional robot simulator Gazebo. The model for the multicopter propeller wash and its aerodynamic interference with the surrounding MAVs is derived. Moreover, the software offers a simplified model of the multiple MAV communication topology.
Besides, the software integrates the embedded flight controller with a finite state machine (FSM) to facilitate MAV deployment and task management. The FSM also supports the use of Simulink model for importing Matlab scripts into the controller. A simulation control program with a graphical interface is designed for Matlab, which enables Matlab to control the Software-in-the-loop simulation.
The proposed system can be applied in the research of MAV flocking consensus algorithm or multi-MAV cooperation, in order to accelerate the development and test, save the experimental costs, and reduce the risks.
Compared with the research of a single MAV, the study of an MAV swarm is more complicated and dangerous. Because multiple MAVs are likely to collide and interfere with each other, their swarm behavior is harder to predict. This paper presents a Software-in-the-Loop (SIL) simulation system for MAV swarming research. This system is able to test the MAV swarm in a software simulated environment.
he SIL simulation system is based on the robot operating system (ROS) and the three-dimensional robot simulator Gazebo. The model for the multicopter propeller wash and its aerodynamic interference with the surrounding MAVs is derived. Moreover, the software offers a simplified model of the multiple MAV communication topology.
Besides, the software integrates the embedded flight controller with a finite state machine (FSM) to facilitate MAV deployment and task management. The FSM also supports the use of Simulink model for importing Matlab scripts into the controller. A simulation control program with a graphical interface is designed for Matlab, which enables Matlab to control the Software-in-the-loop simulation.
The proposed system can be applied in the research of MAV flocking consensus algorithm or multi-MAV cooperation, in order to accelerate the development and test, save the experimental costs, and reduce the risks.