This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Date of publication xxxx 00, 0000, date of current version xxxx 00, 0000.
Digital Object Identifier 10.1109/ACCESS.2017.DOI

Energy Based Control of a Swash Mass

Helicopter through Decoupling Change
of Coordinates
BABAK SALAMAT, AND ANDREA M. TONELLO, (SENIOR MEMBER, IEEE)
University of Klagenfurt, Institute of Networked and Embedded Systems, Chair of Embedded Communication Systems, 9020 Klagenfurt, Austria
Corresponding authors: Babak Salamat (e-mail: [email protected]) and Andrea M. Tonello (e-mail: [email protected]).

ABSTRACT We consider a small helicopter structure that is maneuvered through the control of moving
masses. It is referred to as a swash mass helicopter (SMH). This paper addresses the trajectory tracking
control problem for the SMH, with a specific focus on the decoupling change of coordinates of both
rotational and translational dynamics. We propose a control scheme in which position tracking is the primary
objective, while the attitude tracking task is considered as a secondary objective. The intermediate control
signals related to the attitude dynamics exploit the structural properties of the SMH and are enhanced with
terms that grant a more accurate tracking of the target trajectory. The closed-loop system stability under the
trajectory tracking objective is obtained following the Interconnection and Damping Assignment Passivity-
Based Control (IDA-PBC) approach. In addition, the presence of external disturbances can diminish the
trajectory tracking performance. For this reason, a nonlinear outer loop controller is added to the IDA-
PBC to compensate the disturbances. Finally, the results of several simulations are reported to evaluate the
performance of the control strategy.

INDEX TERMS Unmanned aerial vehicles; swash mass helicopter; decoupling change of coordinates;
passivity-based control; trajectory tracking.

I. INTRODUCTION
HE design of unmanned aerial vehicles (UAVs) has
T gained significant interest due to the large set of ap-
plications ranging from environmental monitoring to packet
Rotor Center (RC)

delivery. Most of the literature focuses on quadrotor heli-

copters [1]–[8]. In this paper, we instead consider a helicopter
structure where maneuvering is obtained by controlling the Geometric Center (GC)
displacement of masses as shown in Fig. 1. In reference to
the UAV structure, the basic idea is to deploy a double blade Swash Mass
coaxial shaft rotor and to maneuver the helicopter through the
linear movement of four masses positioned on the main body ℓ𝑦
plane. We refer to it as the swashed mass helicopter (SMH). Center of Mass (CM)
The rotors provide a drag thrust, while the inertial masses ℓ𝑥
(through the gravitational forces) induce a certain orientation
so that to attain a certain roll, pitch and yaw. Details on the FIGURE 1: Swash mass helicopter.
SMH structure, dynamical system model, and basic control transformed into a canonical form ż◦ = f (z◦ , υ), after ap-
strategy based on the backstepping approach can be found in plying a globally invertible non-linear change of coordinates
[9]. to obtain z◦ = Θ1 (x), υ = Θ2 (x, u) [i.e., x = Θ−1 1 (z0 )],
Basically, from a theoretical point of view, the problem u = Θ−1 2 (x, υ). Here υ is our new control input, and z◦ is
discussed in this paper is an example of a non-linear control used to denote the entire state of our system.
system [10] in the form ẋ = f (x, u) + g(x, u) that can be A solution to the problem can be obtained by decoupling

VOLUME 4, 2016 1

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

both the translational and rotational dynamics [11]. The canonical form and the Euler-Lagrange form. The trajectory
decoupling methodology that we propose is obtained in two tracking problem and the problem of control robustification
steps. The first step is the design of a smooth input moment are formulated and discussed in Section IV. Numerical re-
that decouples the rotational dynamics of our system. In sults are presented in Section V. The conclusions then follow.
the second step, we design a feedback linearization law that Notation. Rn denotes the n-dimensional Euclidean
decouples the translational dynamics of our system. In devel- space and Rm×n the set of m × n real matrices. In and On
oping our methodology we take as well inspiration from [12], are the n × n identity and zeros matrices. For any matrix
in which the emphasis is on defining classes of decoupling A ∈ Rn×n , (Ai ) denotes the i-th column and (Aij ) the
approaches. Moreover, our contribution extends the results ij-th element. The set SO(3) := {R ∈ R3×3 : RT R =
of [12], which only deals with a simple planner aircraft. This I3 , det(R) = 1} denotes the third-order Special Orthogonal
extension is (non-trivially) accomplished via the application group. The vector cross product for vectors a and b in R3 is
of an exact input-output linearization [13]. This decoupling defined as a × b = [a2 b3 − a3 b2 , a3 b1 − a1 b3 , a1 b2 − a2 b1 ].
change of coordinates simplifies the structure of the so-called Given a function f : Rn → R, we define the differential
matching condition and, consequently, it expands the set of operators ∇f := (∂f /∂x)> , ∇xi f := (∂f /∂xi )> , where
target dynamics that can be achieved. xi ∈ Rp is an component of the vector x and k x k2s := x> Sx
With the decoupling methodology the canonical form of where S ∈ Rp×p , S = S > > 0.
the system can be obtained, and consequently the control
mechanism can be designed. Various control methodologies
have been analyzed for UAVs. A linear quadratic regula- II. SYSTEM MODEL
tor (LQR) strategy has been considered in [14], [15] for
The proposed SMH deploys a coaxial double blade rotor with
a quadrotor UAV in the presence of external disturbances.
the rotor connected to the main helicopter body via a rigid
Stochastic feedback control was proposed for a fixed-wing
shaft (main shaft) Fig. 1. To steer the helicopter, four masses
UAV in [16]. From the non-linear control point of view,
are displaced on an orthogonal plane w.r.t. to the rotor shaft
several control laws have been also proposed in recent years
and can be moved with linear cross shaft servos. Assuming
to control the UAV. Back-stepping and sliding mode method-
the rotor and blades to be concentrated in one point in the
ologies were compared in [17]. It was shown that the back-
rotor shaft edge, we refer to such a point as the rotor center
stepping control handles the rotational motion in the presence
(RC). The intersection of the rotor shaft and the swash masses
of large disturbances. A nonlinear geometric control design
plane is referred to as geometrical center (GC) of the SMH.
is developed for a quadrotor in [18]. The Immersion and
The blades rotation induces a thrust aligned with the rotor
Invariance (I&I) approach has been originally presented in
shaft and the swash masses shift the center of mass (CM)
[19] and applied in [20] to deal with the attitude control
of the SMH on an orthogonal plane w.r.t. the thrust vector,
for an octo-copter. Interconnection and damping assignment
i.e.,. it tilts the SMH body. A yaw movement is generated
passivity based control (IDA-PBC) has been applied in [21]–
by changing the relative speed of the two blades since a
[23] for the vertical takeoff and landing of a degree one
drag torque imbalance is generated, while roll and pitch are
under actuated aircraft with strong input coupling. In [24],
regulated by displacing the swash masses asymmetrically
the control of a quadrotor based on the IDA-PBC approach
w.r.t. the GC.
is presented. However, it is limited to tracking of the attitude,
only. In [25], the idea of a quaternion-passivity-based control It is possible to derive the dynamical system model by
is used without the need to solve partial differential equations using the Newton’s framework. We assume the swash masses
(PDEs). to be identical and equal to m. The total SMH mass is given
Considering the SMH, the possibility to include informa- by mt = mb + 4m, where the mass of the structure mb
tion about the underactuated mechanical structure and to deal (not considering the swash masses) is concentrated in the
with the concept of energy shaping in the IDA-PBC control GC (assuming the mass of the rotors in the RC negligible).
design has motivated the work in this paper. Despite the ad- The net thrust generated by the rotors is T1 , while the
vantage of IDA-PBC in stabilization, there is a lack of work gravitational acceleration is g. Details can be found in [9].
about IDA-PBC for trajectory tracking. The contribution in Herein, for compactness, we report only the system model
this paper extends the results of [26], which only deals with result which reads as follows
set-point regulation (constant reference) tasks. In addition,
we show that our control design provides robust tracking by      
ẍ 0 (cφ sθ cψ + sφ sψ )T1
adding a nonlinear outer loop controller. As it will be shown, 1 
ÿ  =  0  + (cφ sθ sψ − sφ cψ )T1  (1)
    
the proposed control design is simple since it does not require mt
solving the nonlinear partial differential equations (PDEs) z̈ −g (cφ cθ )T1
describing the SMH dynamics.
   
φ̈ φ̇ i
The outline of the paper is as follows. The functionality
h
†−1 ˙ † 
 θ̈  = I M − I  θ̇  (2)
 
c
of the SMH structure and the system model are given in
Section II. In Section III, we explain the derivation of the ψ̈ ψ̇

2 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

"  #
with
   Ixy = Iyx = −mb − β`y − β`x
cφ sψ − cψ sφ sθ cψ cθ
sφ sψ + cφ cψ sθ βT1 `y
     
Mc = −cφ cψ − sφ sψ sθ

−cψ sφ + cφ sψ sθ  βT1 `x 
cθ sψ
   1 L  1 L
−m − β `x + − β `y +
−cθ sφ −sθ
cφ cθ Mψ 2 2 2 2
     
(3)  1 L  1 L
−m − β `x − − β `y −
where the terms s. and c. denote the sine and cosine functions 2 2 2 2
of the argument in the subscript, respectively, and I † ∈ R3×3
is the absolute inertia matrix of the system which can be Ixz = Iyz = 0.
expressed as follows:
  The control inputs can be grouped in the control input
I11 I12 I13
vector U = [T1 `y `x Mψ ]T and they are: the thrust T1 , the
I † = I21 I22 I23  . (4)
swash mass position −L ≤ `y ≤ L responsible for the roll
I31 I32 I33
torque, the swash mass position −L ≤ `x ≤ L responsible
The elements of the absolute inertia matrix in (4) can be for the pitch torque, and, finally, the yaw torque Mψ . We
computed as follows assume the two swash masses to be mutually constrained at
constant distance L and the rest position of each masses has
I11 = Ixx
a distance L2 from the center point of the shaft of length L.
I12 = I21 = Ixy cos φ + Ixx sin φ tan θ We define β as the ratio of the swash mass m and the total
m
I13 = I31 = Ixx cos φ tan θ − Ixy sin φ SMH mass mt , i.e., β = m t
.
  The dynamical system model in (1)-(2) is highly nonlinear
I22 = cos φ Iyy cos φ + Ixy sin φ tan θ and shows strong coupling among the rotational and trans-
  Izz sin2 θ lation relations. This renders the development of a control
+ sin φ tan θ Iyx cos φ + Ixx sin φ tan θ − 2 design not straightforward.
 sin θ − 1
I23 = I32 = cos φ tan θ Iyx cos φ + Ixx sin φ tan θ
  Izz cos φ sin θ III. DERIVATION OF THE CANONICAL AND THE
− sin φ Iyy cos φ + Ixy sin φ tan θ + 2 EULER-LAGRANGE FORM
cos θ
In this section, the aim is to decouple the rotational dynamics
  Izz cos2 φ of our system model. Then, we linearize the translational
I33 = sin φ Iyy sin φ − Ixy cos φ tan θ + 2θ
  cos dynamics of the SMH. Finally, the canonical and the Euler-
− cos φ tan θ Iyx sin φ − Ixx cos φ tan θ Lagrange form of the system will be derived.
with
 2 A. DECOUPLING METHODOLOGY FOR THE
Ixx = mb − β`y ROTATIONAL DYNAMICS
  2   2 Basically, our aim is to obtain a representation of (2) in
1 L 1 L canonical form. However, observing (2) we can say that the
+m − β `y + +m − β `y −
2 2 2 2 matrix (3) operating on the control inputs `y , `x and Mψ is
 2 nonsingular and we can decouple the rotational system by
applying the globally defined change of coordinates given
Iyy = mb − β`x
in the following theorem [12]. Since the rotational dynamics
  2   2 in (2) is a fully actuated mechanical system for φ 6= ± π2 ,
1 L 1 L
+m − β `x + +m − β `x − θ 6= ± π2 and, ψ 6= ±π, it is exactly feedback linearizable.
2 2 2 2
Theorem 1: The following change of coordinates:
 2  2     
βT1 `y A1 A2 A3 υ1
Izz = mb − β`y + mb − β`x
βT1 `x  = A4 A5 A6  υ2  (5)
    
2 2
Mψ A7 A8 A9 υ3
   
1 L 1 L
+m − β `x + +m − β `x −
2 2 2 2 decouples the rotational dynamics in (2). The elements Ai
2 2
are given in the Appendix and υ = [υ1 , υ2 , υ3 ]T ∈ R3 is the
   
1 L 1 L
+m − β `y + +m − β `y −
2 2 2 2 new control input vector for the rotational dynamics in (2).
Corollary 1: The global change of coordinates in (5) applied
to the rotational dynamics in (2) automatically gives the
canonical form.
Proof 1: Applying the change of coordinates in (5), one can
VOLUME 4, 2016 3

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

obtain  
φ̇ i      
−1
h ẍ 0 ux
ῡ = I † ˙ † 
υ − I  θ̇  (6) 1  
ÿ = 0 + uy  (16)
   
mt
   
ψ̇ z̈ −g uz
with a new control input ῡ = [ῡ1 , ῡ2 , ῡ3 ]T ∈ R3 . Now,
   
φ̈ ῡ1
based on Theorem 1, we have the canonical form of the
 θ̈  = ῡ2  . (17)
   
rotational dynamics
ψ̈ ῡ3
φ̈ = ῡ1
C. EULER-LAGRANGE FORM
θ̈ = ῡ2
Now, by using (16) and (17) the SMH can be written in the
ψ̈ = ῡ3 . (7) Euler-Lagrange form, as follows
M (q)(q̈) + C(q, q̇)q̇ + ∇V (q) = Gu (18)
B. FEEDBACK LINEARIZATION FOR THE
where the term q = [η1 η2 ]> = [x y z φ θ ψ]> ∈ R6
TRANSLATIONAL DYNAMICS
are the configuration variables, M = M > = I6×6 ∈ R6×6
We begin by considering the translational dynamics in (1) is the inertia matrix and C(q, q̇) = O6×6 ∈ R6×6 denotes
that can be linearized as follows the Coriolis term, u = [ux uy uz ῡ1 ῡ2 ῡ3 ]> ∈ R6 is the
      input vector, matrices ∇V (q) and G are given by
ẍ 0 ux
1  
ÿ  =  0  + uy  (8)
     
mt ∇V (q) = 0 0 −g 0 0 0 (19)
z̈ −g uz  
αI3 O3
G= (20)
where O3 I3
1
where α = mt .
 
ux = cos φ sin θ cos ψ + sin φ sin ψ T1 (9)
  IV. TRAJECTORY TRACKING FOR THE SMH USING
uy = cos φ sin θ sin ψ − sin φ cos ψ T1 (10) IDA-PBC
 
uz = cos φ cos θ T1 , (11) The development of the canonical form using the decoupling
methodology obtained in the previous section, simplifies the
are new control inputs. Note that the desired roll φ∗ and control design. This is because now the SMH can be treated
pitch θ∗ angles are considered as inputs for the rotational as a fully-actuated mechanical system. The main objective
subsystem. Therefore, we can obtain the total thrust T1 , the of the proposed control law is to act on the rotor thrust and
desired roll φ∗ and pitch θ∗ angles as below via the inverse swash masses position so that the SMH can track the desired
transformation of equations (9-11), as follows: target trajectory q ∗ (t) with stable Euler angles.

uz A. PROBLEM FORMULATION
T1 = (12)
cos φ cos θ Let us consider the SMH model in (18). The objective is to
h u sin ψ − u cos ψ i design a control law u = u(t, q, q̇) that ensures
x ∗ y ∗
φ∗ = arcsin (13)
T1
h u cos ψ + u sin ψ i
x ∗ y ∗ lim k q̃(t) k= lim k q(t) − q ∗ (t) k= 0 (21)
θ∗ = arcsin . (14) t→∞ t→∞
T1 cos φ∗
for any arbitrary bounded target time dependent trajectory
The desired yaw angle ψ∗ will be obtained in a way that q ∗ (t) ∈ C 2 .
SMH’s heading and direction of motion in the x-y plane (top
view) are on the same line. Geometrically, the desired yaw B. IDA-PBC METHODOLOGY
angle ψ∗ can be obtained as follows: Now, we can write the Hamiltonian H(q, p) [21] to be the
  summation of the kinetic energy and the potential energy
−1 y∗ − y (total energy), respectively, in the form
ψ∗ = tan . (15)
x∗ − x 1 > −1
p M (q)p + V (q),
H(q, p) = (22)
Having the desired yaw angle ψ∗ , the desired roll φ∗ and 2
pitch θ∗ angle can be easily obtained using (13), and (14), where q ∈ R6 and p ∈ R6 are the generalized po-
respectively. sition and momentum (we have introduced the notation
From the obtained results, we have the canonical form of q = [η1 η2 ]> = [x y z φ θ ψ]> , p = [η̇1 η̇2 ]> =
our full system model [ẋ ẏ ż φ̇ θ̇ ψ̇]> ).
4 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

𝜂2
Planner 𝑞∗ (𝑡) Position Attitude u U Plant
𝑝∗ (𝑡) controller (SMH) 𝜂1
Block controller

𝐏𝐃 +
Feedback
linearization

Canonical form

FIGURE 2: IDA-PBC controller scheme considering of six cascaded PD+ controllers.

Therefore, the SMH dynamical system model can be ifications on the desired potential function in (26), tracking
written in the port-controlled Hamiltonian (PCH) form as can be realized.
follows: Now, the desired PCH model can be written as

∇q Hd
     
∇q H q̇ 0
     
q̇ 0
= [L(q, p) − R(q, p)] + u (23) = [Ld (q, p) − Rd (q, p)] + u (27)
ṗ ∇p H G ṗ ∇p Hd G
with   with
0 I3×3
M −1 Md
 
L(q, p) = , (24) 0
−I3×3 0 Ld (q, p) = −LTd (q, p)
= , (28)
−M −1 Md J (q, p)
 
0 0  
R(q, p) = . (25) 0 0
0 C(q, p) R(q, p) = , (29)
0 BKP B >
Remark 1: From (22) and (23) it can be observed that
where L> d (q, p) is a skew symmetric matrix and KP =
q̇ = M −1 p. In addition, in the following to ease the notation,
KP> > 0 contains design parameters. J (q, p) is left free to
we will omit in some cases the dependency of vectors and
the designer and follows the skew-symmetry property.
matrices on the state q.
Energy shaping and damping injection: The control law
Suppose that [q ∗ (t) p∗ (t)]> is a bounded target trajectory can be written as
and p∗ can be calculated by p∗ = M (q ∗ )q̇ ∗ . Then, the total u = ues + udi . (30)
desired energy function (closed-loop) can be modified in a
more convenient form. In fact, we can transform the closed- The control law in (30) consists of two terms responsible for
loop system model into the port-controlled Hamiltonian form the energy shaping and the damping injection, respectively.
[21]. By equating (23) and (27), one obtains the so-called
Modification of the total energy function: Motivated by matching equations [21]. This leads to the following partial
(22), we propose the desired trajectory dependent Hamilto- differential equations (PDEs):
nian function
G⊥ {∇q H − Md M −1 ∇q Hd + J (q, p)Md−1 p} = 0 (31)
1
Hd (q, p, q ∗ (t), p∗ (t)) = p> Md−1 (q)p + Vd (q) (26) where G⊥ is a full-rank left-annihilator of G.
2
where Vd (q) and Md = Md> > 0 represent the desired
C. CONCLUSIVE RESULTS
closed-loop potential energy function and inertia matrix,
respectively. As already mentioned, the desired potential function can
be considered as a quadratic function plus a gravitational
Notice that the stabilization problem (q, p) = 1
(q ∗ , 0), q ∗ ∈ R6 is a special case of the tracking problem compensator Vd (q) = V (q) + k q − δ(t) k2Kp to ensure
2
(q, p) = (q ∗ (t), p∗ (t)). We will show that with small mod- tracking ability. To do so, we set Md = M and J (q, p) = 0
VOLUME 4, 2016 5

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

and propose the following desired storage function Φ(q, p, ζ) given by [28]
1 > −1 γ1 = q, (37)
Hd (q, p, q ∗ (t), p∗ (t)) = p M (q)p + V (q)
2 γ2 = p + GK2 KI (ζ − α), (38)
1
k q − δ(t) k2Kp
+ (32) γ3 = ζ. (39)
2
where δ(t) = q ∗ (t) + Kp−1 (ṗ∗ + ∇V (q ∗ )). To proceed, we Then, the closed-loop dynamics can be written in trajec-
extend the well-known controller PD+ [27], from an IDA- tory dependent port-controlled Hamiltonian form as follows:
PBC point of view. This is done by starting from (31) and    
obtaining the final controller in the following form γ̇1 06×6 M −1 Md −M −1 GK2
γ̇2  =  −Md M −1 −GKP GT 06×6  ∇Hγ ,
   
T T −1
u = −Kp (q − δ(t)) −Kd M −1 (q)(p − p∗ (t)) γ̇3 K2 G M 06×6 06×6
| {z }| {z } (40)
ues udi
with Hamiltonian
= ∇V (q ) + ṗ − Kp (q − q (t)) − Kd M −1 (q)(p − p∗ (t)),
∗ ∗ ∗
1 1
(33) Hγ (γ) = γ2T Md−1 γ2 + Vd (γ1 ) + kγ3 − αk2KI (41)
2 2
where Kd is a positive definite matrix. The control law in (33) and
with q̇ = M −1 p can be seen as a PD+ controller plus a time ¯
α := (KP KI )−1 d. (42)
dependent shift. The architecture of the controller in (33)
consists of six cascade PD+ controllers where the cascade Proof 2: We differentiate (37) and (38) to obtain
comprises the control of the roll ↔ x directional movement γ̇1 ≡ q̇
and pitch ↔ y directional movement as depicted in Fig. 2.
= M −1 γ2 − M −1 GK2 KI (γ3 − α), (43)
D. DEVELOPMENT OF A CONTROLLER IN THE γ̇2 = ṗ + GK2 KI ζ̇
PRESENCE OF DISTURBANCES = −Md M −1 ∇Vd (q) − GKP GT Md−1 p
Herein, we consider the effect of constant, matched dis- +G(uo + d) + GK2 KI ζ̇
turbances in the port-controlled Hamiltonian system (23). = −Md M −1 ∇Vd (q) − GKP GT Md−1 p + Gd
Let us consider the dynamical system model in (23) with  
the controller in (33). The disturbances propagate and must T T −1 T −1
+G − K2 KI K2 B M ∇Vd − KP B Md GK2 KI ζ
be compensated with the dynamic outer loop control. To
proceed, we consider the system (23) perturbated by an input +GK2 KI K2T GT M −1 ∇Vd |(q,p,ζ)=Θ−1 (γ)
disturbance in closed-loop with the controller un = u + uo , ≡ −Md M −1 ∇Vd (γ1 ) − GKP GT Md−1 γ2 , (44)
leading to the following model and control problem.
Perturbated dynamical system model: Given the pertur- which is the second row of the closed-loop dynamics (40).
bated system Finally, from the last row of (40), we obtian

∇q Hd γ̇3 = K2T GT M −1 ∇Vd (γ1 ) |γ=Θ(q,p,ζ) ≡ ζ̇. (45)

     
q̇ 0 ¯ (34)
= [Ld (q, p)−Rd (q, p)] + (uo + d),
ṗ ∇p Hd G
To prove global tracking of the SMH under disturbances
with d¯ ∈ R6 , we want to find (if possible) a dynamic we take as well inspiration from [29], in which the emphasis
controller uo = β(q, p, ζ), where ζ ∈ R6 is the state of the was on exploiting contraction analysis [30]. To do so, we
controller, that consider the following proposition
Proposition 2: Consider the trajectory dependent port-
• ensures tracking the target trajectory q ∗ (t) for all initial
controlled Hamiltonian in (40). Hγ (γ) is the desired tra-
condition inside the sphere centered at q ∗ (0) and with
jectory dependent Hamiltonian function which satisfies the
radius T , for some ζ∗ ∈ R6 .
following conditions
Proposition 1: Consider the perturbated dynamical system Condition 1:
model in (34) with J = 0 in closed-loop with the PI  
06×6 M −1 Md −M −1 GK2
controller uo = β(q, ζ), with
 −Md M −1 −GKP GT 06×6  ∇Hγ |q∗ (t), p∗ (t), ζ∗
 
β(q, ζ) = −K2 KI K2T GT M −1 ∇Vd (q) − KP KI ζ (35) K2T GT M −1 06×6 06×6
with constant matrices KP > 0, KI > 0 and = q ∗ (t) (46)
K2 := (GT Md−1 G)−1 . (36) Condition 2:
Introduce the globally defined change of coordinates γ = ρI < ∇2 Hγ < βI ∀q ∈ DT (47)
6 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

for positive constants ρ, β and ρ < β. DT is an open subset V. NUMERICAL RESULTS

of R6 which for all t ≥ 0 contains q ∗ (t) and a sphere with In this section, numerical results are reported to evaluate and
constant radius T around q ∗ (t). validate the proposed control strategy. A simulator of the
If conditions (1) and (2) hold true, then the tracking is dynamical system model (1) and (2) in closed-loop with the
granted globally. proposed control law un = u + uo in (33) and (35) has been
Proof 3: Consider (40), q ∗ (t) is a trajectory of this system. developed. Two flying scenarios are considered: rest to rest
We can write the virtual dynamics 1 of (40) as follows: maneuver, and a complex trajectory. The physical parameters
  of the SMH are mt = 1.2 Kg, m = 0.1 Kg, L = 0.2 m,
06×6 M −1 Md −M −1 GK2
d = 0.2 m and g = 9.81 Kgm/s2 . The total time for both
δ q̇ =  −Md M −1 −GKP GT
 2
06×6  ∇ Hγ δq

scenarios is set to Tf = 10s and Tf = 60s, respectively,
K2T GT M −1 06×6 06×6 the sampling time is set to 0.01ms, and the initial conditions
(48) for both scenarios are q0 = [0, 0, 0, 0, 0, 0]> and
To make our notation simple, we define q0 = [0, 0, 0, 0, 0, π3 ]> , respectively.
 
06×6 M −1 Md −M −1 GK2
J ,  −Md M −1 −GKP GT 06×6  (49)
 
K2T GT M −1 06×6 06×6
Now, we need to prove that there exists a constant Φ for
which the corresponding virtual dynamics is contracting. To
proceed, we define
L , ΦJ∇2 Hγ Φ−1 (50)
>
Therefore, it will be shown that L+L is uniformly negative
definite. In detail,
L + L> = ΦJ∇2 Hγ Φ−1 + Φ−> ∇2 Hγ J > Φ>
= (ΦJ + Φ−> )∇2 Hγ (Φ−1 + J > Φ> )
−ΦJ∇2 Hγ J > Φ> − Φ−> ∇2 Hγ Φ−1 . (51)
Now, we consider ∇ Hγ < βI: 2 FIGURE 3: 3D rest to rest maneuver of the swash mass
helicopter with robust IDA-PBC control.
β(ΦJ + Φ−> )(Φ−1 + J > Φ> ) ≥ ΦJ∇2 Hγ J > Φ>
A. REST TO REST MANEUVER
+Φ> ∇2 Hγ Φ−1 + ΦJ∇2 Hγ Φ−1 + Φ−> ∇2 Hγ> J > Φ>
(52) We start by considering a rest to rest maneuver. The SMH
is commanded to fly from a predefined initial to a final
The inequality in (52) with some straightforward calculation position in the 3D space (rest to rest maneuver). To model the
and considering ρI < ∇2 Hγ < βI, yields: trajectory, at least fifth-order polynomial functions should be
β(ΦJ + Φ−1 + Φ−> J > Φ> ) + (β − ρ)ΦJJ > Φ> considered

+(β − ρ)Φ−1 Φ−> ≥ ΦJ∇2 Hγ J > Φ> + Φ> ∇2 Hγ Φ−1 q1∗ (t) = a0 + a1 t + a2 t2 + a3 t3 + a4 t4 + a5 t5 (57)
−1 −>
+ΦJ∇ Hγ Φ 2
+Φ ∇2
Hγ> J > Φ> . (53) with the six boundary conditions
In fact, (49) is Hurwitz, so there exists a positive definite q1 (t0 ) = 0 q1 (Tf ) = 5 (58)
matrix P such that the following algebraic Riccati equation q̇1 (t0 ) = 0 q̇1 (Tf ) = 0 (59)
is satisfied:
q̈1 (t0 ) = 0 q̈1 (Tf ) = 0 (60)
P J + J > P = −P JJ > P − I. (54)
1
In the considered case, the coefficients of the desired polyno-
Pre- and post-multiplying of (54) by P − 2 gives: mial in (57) are
1 1 1 1 1 1
P 2 JP − 2 + P − 2 J > P 2 = −P − 2 JJ > P 2 − P −1 . (55) a0 = a1 = a2 = 0
Substituting Φ = P
1
and considering (51)-(53), we obtain 1 h i
2
a3 = 20q1 (Tf )
1 1
2Tf3
L + L> ≤ −ρP 2 JJ > P 2 − ρP −1 . (56) 1 h i
a4 = 4 − 30q1 (Tf )
Now, we can state that the right-hand side of the inequality in 2Tf
(56) is negative definite, which completes the proof. 1 h i
a5 = 5 12q1 (Tf ) . (61)
1 For more details and the generalization of the proposed theory see [29].
2Tf

VOLUME 4, 2016 7

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

Therefore, the desired trajectory for the x direction q1∗ (t) = 18 10-3
∗ 1
x (t) is given by
0.5
16
q1∗ (t) = a3 t3 + a4 t4 + a5 t5 . (62)

T1 [N]

lx [m]
0
14
∗ ∗ ∗ ∗ ∗
Since we have defined q (t) = [x (t), y (t), z (t), φ (t) , -0.5

θ∗ (t) , ψ ∗ (t)]> , the desired trajectories for q2∗ (t) = y ∗ (t) and 12
0 2 4 6 8 10
-1
0 2 4 6 8 10
q3∗ (t) = z ∗ (t) directions are given by similar derivations. Time [s] Time [s]
10-3
The proposed desired trajectory (57) with coefficients (61) 1 0.01

ensures that the velocities and accelerations at the rest point 0.5
0

M [Nm]
are zero.

ly [m]
0
Remark 2: It should be noted that the desired roll φ∗ and -0.5
-0.01

pitch θ∗ angles can be obtained by the inverse kinematic in

-1 -0.02
(13) and (14), respectively. 0 2 4 6 8 10 0 2 4 6 8 10
Time [s] Time [s]
The control gains (33) and (35) for this scenario are
  FIGURE 5: Time history of the control inputs.
1 0 0 0 0 0
0 1 0 0 0 0 B. COMPLEX TRAJECTORY
 
0 0 6 0 0 0
Kp =   , We now consider a complex trajectory. The SMH starts at
0 0 0 12 0 0 rest with initial configuration q0 = [0, 0, 0, 0, 0, π3 ]> .

0 0 0 0 12 0
The control gains (33) and (35) are
0 0 0 0 0 3  
  0.8 0 0 0 0 0
2.5 0 0 0 0 0  0 0.6 0 0 0
 0 2.7 0 0 0 0   0
  0 0 5 0 0 0
0 0 3.5 0 0 0  Kp =   ,
Kd =  0
,
0 0 0 16 0 0
0 0 6 0 0 0


0
 0 0 0 16 0 
0 0 0 6 0
0 0 0 0 0 1.9
0 0 0 0 0 1.5  
3.8 0 0 0 0 0
KI = I6 .  0 3.8 0
 0 0 0 
In Fig. 3, we report the details of the trajectory obtained from 0 0 1.5 0 0 0
Kd =   ,
the control law un = u + uo when the SMH is commanded 0 0 0 10 0 0 
to follow the rest to rest maneuver in the 3D space. The 0 0 0 0 10 0
details of the 3D trajectory is shown in Fig. 4. Based on the 0 0 0 0 0 2
results shown in Fig. 4, it can be concluded that the transient K I = I6 .
performance is quite good and the SMH can reach the stable
position in a short time. Finally, the control inputs are shown The considered target trajectory for the SMH is
in Fig. 5.
      π 
q1∗ (t) x∗ (t) 2 1 − cos( t)
10
Desired position
10
Desired position
 ∗   ∗  
q2 (t) = y (t) =  π 18 
,
Robust IDA-PBC 2 sin( t)
x [m]

Robust IDA-PBC
y [m]

9
 
5 5
q3∗ (t) z ∗ (t) t
0
0 2 4 6 8 10
0
0 2 4 6 8 10 q6∗ (t) = ψ ∗ (t) = 0 rad.
Time [s] Time [s]
10 4
Desired altitude Desired pitch angle
In Fig. 6, we report the details of the trajectory obtained
2
[deg]

Robust IDA-PBC Robust IDA-PBC from the control law un = u + uo when the SMH is
z [m]

5 0
-2 commanded to follow the given target trajectory in the 3D
0
0 2 4 6 8 10
-4
0 2 4 6 8 10
space. To show the effectiveness of the proposed controller
Time [s] Time [s] in the presence of a disturbance, we add a matched constant
4
data1
60
Desired yaw angle
disturbance d¯ = 0.8 at time t = 20s (Fig. 6) to the system.
2 40
The details of the 3D trajectory are shown in Fig. 7. The blue,
[deg]

Robust IDA-PBC Robust IDA-PBC

[deg]

20
0
-2
0 green dashed, and red lines represent the target trajectory,
-20
-4 -40 and the ones obtained with the IDA-PBC controller and, the
0 2 4 6 8 10 0 2 4 6 8 10
Time [s] Time [s] robust IDA-PBC controller, respectively. Fig. 7 shows the
tracking performance of the translational position, x, y, and
FIGURE 4: Trajectory time evolution of the SMH with the z, and the yaw angle ψ, and the stabilization of the roll and
robust IDA-PBC controller. pitch angles, φ and θ, respectively. As expected, the presence
8 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

TABLE 1: Root-mean-square error between desired and real

trajectory in meters. in the control signal. Finally, the control inputs are shown
in Fig. 9. On the other hand, during the initial states, when
Controller RMSE(x) RMSE(y) RMSE(z) the desired attitude is not achievable, the SMH is working
Robust IDA-PBC 0.2022 0.1410 0.0604 and moving in an energy-efficient way in the sense of Propo-
IDA-PBC 1.3338 0.2719 0.0950 sition 2 to damp and modify the attitude reference so that
stabilization is guaranteed.

of the disturbance produces an error in tracking, so that the 2 2

SMH diverges from the target path. However, by adding the

and d
and d
1 1
PI controller we can reject the disturbance. 0 0

2
1
Despite the aggressiveness of the target trajectory, the -1 -1
0 20 40 60 0 20 40 60
SMH is able to well follow it. The real-target trajectory Time [s] Time [s]
1
overall RMSE is reported in Table 1. 1

and d
and d
0.5 0.5

0 0

4
3
-0.5 -0.5
0 20 40 60 0 20 40 60
Time [s] Time [s]
1 2

and d

and d
0.5
0
0

6
-0.5 -2
0 20 40 60 0 20 40 60
Time [s] Time [s]

FIGURE 8: Time history of the matched disturbance d, and

the controller state ζ.

20 0.01
Robust IDA-PBC Robust IDA-PBC
18 IDA-PBC IDA-PBC
0.005
16
T1 [N]

lx [m]
0
14

FIGURE 6: 3D trajectory tracking of the swash mass heli- 12 -0.005

copter with the IDA-PBC and robust IDA-PBC controllers. 10 -0.01
0 20 40 60 0 20 40 60
Time [s] Time [s]
8 4
6 0.01 0.01
2
Robust IDA-PBC Robust IDA-PBC
y [m]
x [m]

4
0 IDA-PBC IDA-PBC
2 0.005 0.005
-2
M [Nm]

0
ly [m]

-2 -4 0 0
0 20 40 60 0 20 40 60
Time [s] Time [s] -0.005 -0.005
60 10
Desired trajectory
[deg]

40 Robust IDA-PBC -0.01 -0.01

z [m]

IDA-PBC 0 0 20 40 60 0 20 40 60
20 Time [s] Time [s]

0 -10
0 20 40 60 0 20 40 60
Time [s] Time [s] FIGURE 9: Time history of the control inputs.
10 60
40
[deg]

[deg]

0
20 VI. CONCLUSION
0
-20 We have briefly presented a novel small aerial vehicle struc-
-10
0 20 40 60
-40
0 20 40 60
ture, referred to as swash mass helicopter (SMH) that allows
Time [s] Time [s] maneuvering it through the control of four swash masses
and a double coaxial rotor. We have then focused the at-
FIGURE 7: Time histories of the SMH with the IDA-PBC and tention to the design of an automatic control mechanism
robust IDA-PBC controller.
so that the SMH tracks a target trajectory with stable Euler
Fig. 8 shows the time history of the controller state ζ, angles. The dynamical system of equations that describes the
which provides the nice disturbance rejection. As can be SMH dynamics is non-linear and rather unique. In fact, it
seen, the controller can track the disturbance and reject it consists of sub-sets of differential equations highly-coupled
perfectly. The proposed robust control law based on IDA- through the control inputs. Considering the full 3D case,
PBC does not include the second time derivatives of the we essentially highlighted that a key point for the control
desired states. This is of great advantage because in practice, of the SMH is the ability to decouple its three second-
the numerical computation of the derivatives generates noise order subsystems using a global change of coordinates. This
VOLUME 4, 2016 9

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

allows us to transform the dynamical system model into the [14] B. Salamat and A. M. Tonello, “Novel trajectory generation and adaptive
canonical form so that an IDA-PBC control approach can evolutionary feedback controller for quadrotors,” in 2018 IEEE Aerospace
Conference Big Sky, Montana, USA, March 2018.
be applied. The control laws have been derived including [15] M. Faessler, D. Falanga, and D. Scaramuzza, “Thrust mixing, saturation,
a modified version to cope with external disturbances. An and body-rate control for accurate aggressive quadrotor flight,” IEEE
analysis of robustness exploiting contraction theory has also Robotics and Automation Letters, vol. 2, no. 2, pp. 476–482, April 2017.
[16] R. P. Anderson and D. Milutinović, “A stochastic approach to dubins
been developed. Several results from simulations have been vehicle tracking problems,” IEEE Transactions on Automatic Control,
presented to assess the performance of the SMH control. vol. 59, no. 10, pp. 2801–2806, Oct 2014.
They show that good controllability of the SMH is attainable. [17] S. Bouabdallah and R. Siegwart, “Backstepping and sliding-mode tech-
niques applied to an indoor micro quadrotor,” in Proceedings of the 2005
IEEE International Conference on Robotics and Automation, April 2005,
APPENDIX. THE DECOUPLING MATRIX pp. 2247–2252.
[18] M. Ryll, H. H. Bülthoff, and P. R. Giordano, “Modeling and control
The elements Ai , i = 1, ..., 9, in (5) are of a quadrotor uav with tilting propellers,” in 2012 IEEE International
Conference on Robotics and Automation, May 2012, pp. 4606–4613.
A1 = cos(φ) sin(ψ) − cos(ψ) sin(φ) sin(θ) [19] A. Astolfi and R. Ortega, “Immersion and invariance: a new tool for stabi-
lization and adaptive control of nonlinear systems,” IEEE Transactions on
A2 = cos(ψ) cos(θ) Automatic Control, vol. 48, no. 4, pp. 590–606, April 2003.
A3 = sin(φ) sin(ψ) + cos(φ) cos(ψ) sin(θ) [20] Y. Bouzid, H. Siguerdidjane, and Y. Bestaoui, “Hierarchical Autopilot
Design based on Immersion & Invariance and Nonlinear Internal Model
A4 = cos(φ) cos(ψ) + sin(φ) sin(ψ) cos(θ) Tracking Controllers for Autonomous system,” in 4th IFAC International
Conference on Intelligent Control and Automation Sciences (ICONS
A5 = cos(θ) sin(ψ) 2016), ser. IFAC-PapersOnLine, vol. 49, no. 5, Reims, France, 2016, pp.
A6 = − cos(ψ) sin(φ) + cos(φ) sin(ψ) sin(θ) 103–108.
[21] J. A. Acosta, R. Ortega, A. Astolfi, and A. D. Mahindrakar, “Interconnec-
A7 = cos(θ) sin(φ) tion and damping assignment passivity-based control of mechanical sys-
tems with underactuation degree one,” IEEE Transactions on Automatic
A8 = sin(θ) Control, vol. 50, no. 12, pp. 1936–1955, Dec 2005.
[22] R. Ortega, M. W. Spong, F. Gomez-Estern, and G. Blankenstein, “Stabi-
A9 = cos(φ) cos(θ).
lization of a class of underactuated mechanical systems via interconnection
and damping assignment,” IEEE Transactions on Automatic Control,
vol. 47, no. 8, pp. 1218–1233, Aug 2002.
REFERENCES [23] C. Venkatesh, R. Mehra, F. Kazi, and N. M. Singh, “Passivity based
[1] B. Salamat and A. M. Tonello, “Adaptive nonlinear PID control for controller for underactuated pvtol system,” in 2013 IEEE International
a quadrotor UAV using particle swarm optimization,” in 2019 IEEE Conference on Electronics, Computing and Communication Technologies,
Aerospace Conference Big Sky, Montana, USA, March 2019. Jan 2013, pp. 1–5.
[2] G. Antonelli, E. Cataldi, F. Arrichiello, P. Robuffo Giordano, S. Chi- [24] B. Yüksel, C. Secchi, H. H. Bülthoff, and A. Franchi, “Reshaping the
averini, and A. Franchi, “Adaptive trajectory tracking for quadrotor mavs physical properties of a quadrotor through ida-pbc and its application to
in presence of parameter uncertainties and external disturbances,” IEEE aerial physical interaction,” in 2014 IEEE International Conference on
Transactions on Control Systems Technology, vol. 26, no. 1, pp. 248–254, Robotics and Automation (ICRA), May 2014, pp. 6258–6265.
Jan 2018. [25] M. E. Guerrero-Sanchez, H. Abaunza, P. Castillo, R. Lozano, C. Garcia-
[3] B. Zhao, B. Xian, Y. Zhang, and X. Zhang, “Nonlinear robust adaptive Beltran, and A. Rodriguez-Palacios, “Passivity-based control for a micro
tracking control of a quadrotor uav via immersion and invariance method- air vehicle using unit quaternions,” Applied Sciences, vol. 7, no. 1, 2017.
ology,” IEEE Transactions on Industrial Electronics, vol. 62, no. 5, pp. [Online]. Available: https://www.mdpi.com/2076-3417/7/1/13
2891–2902, May 2015. [26] B. Salamat and A. M. Tonello, “Robust energy-based control of a swash
[4] D. J. Almakhles, “Robust backstepping sliding mode control for a quadro- mass helicopter subject to matched disturbances,” in Proc. of IEEE
tor trajectory tracking application,” IEEE Access, vol. 8, pp. 5515–5525, Aerospace 2018, Yellowstone, USA, March 7-14, March 2020.
2020. [27] B. Paden and R. Panja, “Globally asymptotically stable ‘pd+’ controller
for robot manipulators,” International Journal of Control, vol. 47, no. 6,
[5] A. C. Satici, H. Poonawala, and M. W. Spong, “Robust optimal control of
pp. 1697–1712, 1988.
quadrotor uavs,” IEEE Access, vol. 1, pp. 79–93, 2013.
[28] A. Donaire, J. G. Romero, R. Ortega, B. Siciliano, and M. Crespo, “Robust
[6] B. Tian, Y. Ma, and Q. Zong, “A continuous finite-time output feedback
IDA-PBC for underactuated mechanical systems subject to matched dis-
control scheme and its application in quadrotor uavs,” IEEE Access, vol. 6,
turbances,” International Journal of Robust and Nonlinear Control, vol. 27,
pp. 19 807–19 813, 2018.
no. 6, pp. 1000–1016, 2017.
[7] Z. Liu, L. Xiong, and W. Shi, “Trajectory tracking control for a quav [29] A. Yaghmaei and M. J. Yazdanpanah, “Trajectory tracking of a class of
with performance constraints,” IEEE Access, vol. 7, pp. 142 467–142 477, port hamiltonian systems using timed ida-pbc technique,” in 2015 54th
2019. IEEE Conference on Decision and Control (CDC), Dec 2015, pp. 5037–
[8] J. Raj, K. S. Raghuwaiya, and J. Vanualailai, “Novel lyapunov-based 5042.
autonomous controllers for quadrotors,” IEEE Access, vol. 8, pp. 47 393– [30] W. Lohmiller and J.-J. E. S. *, “Contraction analysis of non-linear dis-
47 406, 2020. tributed systems,” International Journal of Control, vol. 78, no. 9, pp. 678–
[9] A. M. Tonello and B. Salamat, “A Swash Mass Unmanned Aerial Vehicle: 688, 2005.
Design, Modeling and Control,” arXiv e-prints, p. arXiv:1909.06154, Sep
2019.
[10] R. Olfati-Saber, “Global configuration stabilization for the vtol aircraft
with strong input coupling,” IEEE Transactions on Automatic Control,
vol. 47, no. 11, pp. 1949–1952, Nov 2002.
[11] R. Olfati-Saber, “Nonlinear control of underactuated mechanical systems
with application to robotics and aerospace vehicles,” Ph.D. dissertation,
Massachusetts Institute of Technology, 2001.
[12] S. Sastry, Nonlinear Systems: Analysis, Stability, and Control, ser. Inter-
disciplinary Applied Mathematics. Springer New York, 1999.
[13] H. K. Khalil, Nonlinear systems; 3rd ed. Upper Saddle River, NJ:
Prentice-Hall, 2002.

10 VOLUME 4, 2016

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.
 This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication. Citation information: DOI
10.1109/ACCESS.2020.2990113, IEEE Access

Author et al.: Preparation of Papers for IEEE TRANSACTIONS and JOURNALS

BABAK SALAMAT received the B.S. in mechani-

cal engineering and M.S. degrees in aerospace en-
gineering from the Air-force University of Shahid
Sattari, Tehran-Iran, in 2012 and 2014, respec-
tively. He is currently a Ph.D. student in infor-
mation technology at the University of Klagen-
furt, Austria. His research interest includes path
planning, optimal control, nonlinear and energy-
based control theory with application to robotics
and aerospace systems. He is the co-recipient with
A. Tonello of the 2018 best paper award in the Aerospace journal.

ANDREA M. TONELLO received the D.Eng. de-

gree (Hons.) in electronics and the D.Res. degree
in electronics and telecommunications from the
University of Padova, Italy, in 1996 and 2002,
respectively. From 1997 to 2002, he was with Bell
Labs-Lucent Technologies, Whippany, NJ, USA,
as a member of the Technical Staff. Then, he was
promoted to Technical Manager and appointed to
Managing Director of the Bell Labs Italy Division.
In 2003, he joined the University of Udine, Udine,
Italy, where he became an Aggregate Professor in 2005 and an Associate
Professor in 2014. He is currently the Chair of the Embedded Communi-
cation Systems Group, University of Klagenfurt, Klagenfurt, Austria. He
is also the Founder of the spinoff company, WiTiKee. He received several
awards, including the Distinguished Visiting Fellowship from the Royal
Academy of Engineering, U.K. (2010), the IEEE VTS and COMSOC
Distinguished Lecturer Awards (2011, 2015, 2018), the Chair of Excellence
from UC3M (2019-20), and nine best paper awards. He served/s as an
Associate Editor of the IEEE Trans. on Vehicular Techn., IEEE Trans. on
Communications, IEEE Access, and IET Smart Grid. He served as the Chair
of the IEEE ComSoc TC-PLC. He currently serves as the Chair of the IEEE
ComSoc TC-SGC and he has been appointed Director of Industry Outreach
within IEEE ComSoc.

VOLUME 4, 2016 11

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/.