Multi-vehicle dynamics and control for aerial recovery of micro air vehicles

Abstract

Aerial recovery of micro air vehicles (MAVs) presents a challenging problem in multi-vehicle dynamics and control. This paper presents a method for recovering MAVs in ight using a mothership and towed drogue, in which the mothership executes an orbit that places the drogue in a stable, slower orbit that can be tracked by a MAV. A method for modeling the dynamics of the mothership-cabledrogue system, based on Gauss's principle, is presented. The dierential atness property of the system is exploited to calculate mothership trajectories from desired drogue orbits, and a backstepping controller is proposed that enables accurate mothership trajectory tracking. A drag-based controller for the drogue is also described. Methods to enable the MAV to estimate and track the drogue orbit are discussed. The modeling and control methods are illustrated through simulation and ight results.

Parameter estimation for towed cable systems using moving horizon estimation

“…The work related to the aerial recovery project in which a towed cable system was used to retrieve Miniature Air Vehicles (MAVs) can be found in [16,[18][19][20][21][22][23]. The basic concept of the aerial recovery mission is described in Figure 1, where the towing vehicle (mothership) is commanded to follow a circular path that results in an orbit of the towed body (drogue) with smaller radius and lower speed relative to the mothership.…”

“…The baseline concept of aerial recovery using a towed cable system. studies for the aerial recovery system include mathematical modeling approaches using discrete models [18,21], motion planning and control strategies [19,20,22,23], and the experimental validation of both the mathematical model [16] and control strategies [22,23] using flight tests.…”

“…In the literature, Lagrange's method [4,30], Kane's equations [15,31], Gauss's Principle [18,21], and Newton's second law [19,20] have been applied to establish the equations of motion for towed cable systems. Lagrange's and Kane's approaches require explicit description of internal and external forces and do not express the system dynamics in a concise way for a large number of cable links.…”

Optimal Trajectory Generation Using Model Predictive Control for Aerially Towed Cable Systems

“…In this paper we focus on optimal, open-loop trajectory design for the towing vehicle to place the drogue into a specified orbit. Additional work on various aspects of this problem can be found in [1][2][3][4][5][6][7][8]. The system shown in Fig.…”

“…Existing methods for generating the desired trajectory for the mothership can be classified into two categories: differential flatness based methods [1,3,6,19,27] and optimal control based methods [14,17,21,[28][29][30].…”