Abstract:
In this paper, a discrete-time method for the formation of high-altitude balloons is developed. The balloons float passively along the Earth's wind currents. For actuatio...Show MoreMetadata
Abstract:
In this paper, a discrete-time method for the formation of high-altitude balloons is developed. The balloons float passively along the Earth's wind currents. For actuation, a balloon can change its altitude to enter a different wind current and move in a different direction. The control objective is to steer a fleet of balloons into a configuration where they are evenly distributed around the Earth. The control approach is a discrete-time distributed extremum-seeking controller. This controller works to minimize a measured cost function. It only requires a measurement of this cost function and does not require a model of the nonlinear time-varying wind currents. For the problem of balloon formation control, the cost function is based on a Voronoi partition resulting in an algorithm similar to Lloyd's algorithm. The control architecture is fully distributed. There is no central coordinator and each balloon receives all the information it needs by communicating to nearby balloons over a network whose structure is the Delaunay triangulation with the balloons as vertices. The resulting distributed control algorithm is computationally efficient as the burden of computation is shared between all of the balloons. Several simulations involving 1200 balloons are used to verify the effectiveness of this approach. The simulations use realistic nonlinear time-varying models which are obtained by interpolating gridded weather data obtained from the National Oceanic and Atmospheric Administration.
Published in: 2017 American Control Conference (ACC)
Date of Conference: 24-26 May 2017
Date Added to IEEE Xplore: 03 July 2017
ISBN Information:
Electronic ISSN: 2378-5861