Cooperative Control Reconfiguration in Multiple Quadrotor Systems with
Actuator Faults
Huiliao Yang. Bin Jiang. Hao Yang. Ke Zhang
College of Automation Engineering, Nanjing University of Aeronautics and Astronautics,
Nanjing, China.
Abstract: A cooperative control problem for a multi-quadrotor system with quadrotor actuator faults is
investigated. In the fault-free situation, distributed dynamic surface control is employed to design the
nonlinear cooperative controllers, under which the team of quadrotors can move together in a prescribed
leader-follower formation to reach a common target point. Once the actuators of some quadrotors become
faulty, fault detection and isolation (FDI) detects the faulty quadrotors which are forced to leave the
formation safely. Meanwhile, formation reconfiguration is constructed to make the remaining healthy
quadrotors select another formation to reach the target point. The collision avoidance is achieved in the
whole formation reconfiguration process by using a potential function. Simulation results show the
efficiency of the proposed method.
Keywords: Cooperative Control, Quadrotor, Actuator Fault, Formation Reconfiguration.
1. INTRODUCTION
Quadrotor is one kind of electric vertical take-off and landing
(VTOL) helicopters. Compared with the conventional rotor
helicopter, it can generate more lift force and has more
compact structure. Especially, the four rotors can counteract
the reaction torque mutually, therefore, the propellers against
reaction torque are not needed as in Nie et al. (2007). Due to
these properties, quadrotor has broader military and civilian
prospect, such as detecting, monitoring and surveillance.
However, to complete the above mentioned tasks
successfully, multi-quadrotor system becomes a promising
solution, because it can provide the flexibility in performing
the task and making the system more tolerant to possible
individual quadrotor faults. Moreover, a group of small
quadrotors is more economic than a complex big one (Zhang
and Mehrjerdi (2013)). To operate this team, formation flight
control is considered as one key technique, which contains
quadrotor control and cooperative control.
One of important research areas of cooperative control is
fault tolerance. If some members in the formation have
actuator faults, they may deviate from the desired path, or
even collide with other healthy individuals. Hence, effective
cooperative control which forces the faulty quadrotors to
leave the whole formation safely and formation
reconfiguration are of necessity in practice.
Research supported by Natural Science Foundation of China (61273171,
61473143, 61304112), Fundamental Research Funds for the Central
Universities (NE2014202).
Corresponding author Bin Jiang, Tel: +86 25 84892305-6041. Email
addresses: yhl19901202@126.com (Huiliao Yang), binjiang@nuaa.edu.cn
(Bin Jiang), haoyang@nuaa.edu.cn (Hao Yang), kezhang@nuaa.edu.cn (Ke
Zhang).
On the issues of formation reconfiguration, the previous
works mostly centralize on obstacle avoidance problem in
Murray and Saber (2003), Duan et al. (2008), Lie and Go
(2013), rarely involve the reconfiguration under individual
fault. In Boskovic and Mehra (2002), the case of over-
actuated UAV is considered. An adaptive reconfigurable
formation control utilizing multiple models when the
actuators of the follower vehicle are lock-in-place is put
forward. However, the stability of the whole multiple models
based system cannot be guaranteed. In Longhi et al. (2008), a
cooperative control of multi-underwater vehicle system is
investigated. Based on Decentralized Model Predictive
Control (DMPC) and FDI strategy, the fleet can allow
cooperation and manage the faulty situations in which
reorganizing their formation and rerouting the information
flow are needed. But how the faulty vehicles leave the
formation safely is not considered. In Gallehdari et al. (2014),
a reconfigurable control protocol for a linear multi-agent
system is designed to seek consensus in the presence of
actuator faults and saturations.
In this paper, we consider the lead-follower target
aggregation problem for a multi-quadrotor system with
actuator faults on individual vehicles. In fault-free situation,
dynamic surface control algorithm, the advantage of which is
overcoming the problem of “explosion of terms” in the
backstepping method (Swaroop et al. (2000)), can ensure
three dimensional positions and the heading directions of all
agents to reach a common target point, namely, one kind of
target aggregation problems. When there are actuator faults,
such as losses of control effectiveness or floats, the
decentralized FDI works to find the faulty members which
will escape from this formation. In the escape maneuver, the
potential-like function based controllers are used to prevent
the leaving ones from colliding with others. On the other
Preprints, 9th IFAC Symposium on Fault Detection, Supervision and
Safety of Technical Processes
September 2-4, 2015. Arts et Métiers ParisTech, Paris, France
Copyright © 2015 IFAC 386