electronics

Article
Battery Energy Management of Autonomous Electric
Vehicles Using Computationally Inexpensive Model
Predictive Control
Kyoungseok Han 1 , Tam W. Nguyen 2 and Kanghyun Nam 3, *
1 School of Mechanical Engineering, Kyungpook National University, Daegu 41566, Korea;
[email protected]
2 Department of Aerospace Engineering, University of Michigan, Ann Arbor, MI 48109, USA;
[email protected]
3 Department of Mechanical Engineering, Yeungnam University, Gyeongsan 38541, Korea
* Correspondence: [email protected]; Tel.: +82-53-810-2455




Received: 14 July 2020; Accepted: 5 August 2020; Published: 9 August 2020


Abstract: With the emergence of vehicle-communication technologies, many researchers have

strongly focused their interest in vehicle energy-efficiency control using this connectivity. For instance,
the exploitation of preview traffic enables the vehicle to plan its speed and position trajectories
given a prediction horizon so that energy consumption is minimized. To handle the strong
uncertainties in the traffic model in the future, a constrained controller is generally employed
in the existing researches. However, its expensive computational feature largely prevents its
commercialization. This paper addresses computational burden of the constrained controller by
proposing a computationally tractable model prediction control (MPC) for real-time implementation in
autonomous electric vehicles. We present several remedies to achieve a computationally manageable
constrained control, and analyze its real-time computation feasibility and effectiveness in various
driving conditions. In particular, both warmstarting and move-blocking methods could relax the
computations significantly. Through the validations, we confirm the effectiveness of the proposed
approach while maintaining good performance compared to other alternative schemes.

Keywords: self-driving car; model predictive control; dynamic programming; prediction horizon;
move-blocking; warmstarting

1. Introduction
For the last few years, many experts in the field of automotive research have claimed that
self-driving cars will become commonplace in the near future. Although there are still technical
difficulties for commercializing self-driving cars, at least a certain degree of automation in the vehicles
is expected to be available soon [1,2].
With the advent of new features in self-driving cars, novel approaches to optimize the
energy efficiency of vehicles have become available. In general, human-driver behaviors are
based on their own driving habits on the roads, which are not consistent with the traffic flow [3].
Self-driving cars, in contrast, can plan optimally their trajectories by exploiting real-time traffic
information [4–6]. For example, many short and long-range traffic information using vehicle-to-vehicle
(V2V), vehicle-to-infrastructure (V2I), and, more generally, vehicle-to-everything (V2X) communication
(see Figure 1) provide useful traffic information, which the self-driving cars can effectively use in real
time. For example, the geometry information of the route is available through the pre-loaded maps
and the future traffic density is also available through the V2X technologies. Therefore, unlike human

Electronics 2020, 9, 1277; doi:10.3390/electronics9081277 www.mdpi.com/journal/electronics

Electronics 2020, 9, 1277 2 of 19

drivers, self-driving cars can make suitable decisions based on real-time traffic flow so as to optimize
their energy efficiency.

Figure 1. Increased levels of automation and connectivity through vehicle-to-everything (V2X)

communication.

However, since the traffic dynamics are complex and nonlinear, there is a non-negligible
computational burden associated with long-term real-time traffic prediction. Moreover, since the
traffic predicted by the self-driving car in the long run become uncertain (e.g., traffic disturbances,
unexpected changes in human-driver behavior), it is required to choose an appropriate length of the
prediction horizon to effectively exploit the current traffic information.
Although there have been numerous researches that show the benefits of vehicle speed- and
position-trajectory planning in terms of energy efficiency, to the best of our knowledge, practical
aspects, such as the length of the prediction horizon and its relative computational burden, have not
been discussed in the existing literature on self-driving cars. In order to achieve high performance
while ensuring computational feasibility, this paper deals with the above-mentioned practical concerns
for real production vehicles.
Also, in this paper, the battery-electric vehicle (BEV) is assumed to be ego vehicle that needs to be
controlled, so, using connectivity technologies (i.e., V2X), the optimal battery energy management is
our ultimate goal.

1.1. Literature Review

So far, there have been many research interests in the optimization of energy efficiency using
constrained control theories. In general, both vehicle- and powertrain-level optimizations are used
to achieve specific control objectives while ensuring constraint satisfaction. For example, the vehicle
speed and position are planned in a way that minimizes the energy consumption, while at the same
time the operating points in the powertrain such as engine, motor, and transmission are moved to the
optimum area.
In particular, the BEV that is entirely powered by the electric motors is promising platform
for the autonomous vehicles due to its environmental benefits and low operating cost. In [7],
as compared to the internal-combustion engine vehicle, the benefit of BEV in terms of energy
requirement per unit distance has been confirmed when both powertrain types were utilized as
the fully autonomous vehicles. In addition, the synergies between shared autonomous vehicle fleets
and electric vehicle has also been proposed in [8]. For this reason, this paper also employs the BEV as
our vehicle platform and the minimization of battery energy consumption of BEV has been pursued.
To achieve this goal, one representative method is to adopt dynamic programming (DP),
which ensures global optimality by exploring all possible combinations of the trajectories. DP requires,
however, the entire traffic prediction before the trip. In [9], global optimum speed trajectory of the
vehicle could be obtained using DP, but future traffic environments even including pre-known of the
leading vehicle’s speed profiles are assumed to be given initially. That is, DP is mainly suitable for
offline computation and is thus not implementable in real time due to the heavy computational burden.
However, the control performance using DP is meaningful as a benchmark to investigate the maximum
achievable control performance.
Electronics 2020, 9, 1277 3 of 19

To manage the computational complexity of DP, another well-known trajectory optimization

approach is Pontryagin Minimum Principle (PMP), which has been utilized in [10], for example.
Note that only the necessary condition for optimality is guaranteed, and the method still requires a
long-term prediction. Moreover, two-point boundary value problem associated with PMP conditions
cannot be easily solved numerically. Unfortunately, most literature that employ the PMP [10,11] do
not handle the tractability of their optimal controller rigorously.
Although these two approaches, i.e., DP and PMP, have shown the effectiveness when
planning the optimal trajectory, these methods are usually effective for the deterministic system.
However, the interactions between the road participants in the traffic cannot be modelled in
deterministic way. To handle the such uncertainties in the traffic prediction, stochastic DP [12],
Q-learning-based PMP [13], and reinforcement learning [14,15] have been considered, but the analysis
on the practical aspects like computational tractability is omitted. Additionally, since these methods are
based on Markov decision process, where the relationship between the control action and probability
are pre-determined, these methods cannot ensure robustness when the vehicle is in an unexpected
situation [16].
The above-mentioned two approaches, that is, DP and PMP, are the most popular methods and
provide an optimal-state-trajectory framework. However, as discussed, the heavy computational
burden prevents the introduction of such techniques in production vehicles. Also, the requirement
of long-term preview information in DP and PMP is not usually available in the real-world
traffic since the traffic is significantly influenced by the unexpected uncertainties. To address this,
the method that re-calculate the optimal trajectory once the new traffic information become available
can be considered. More appropriately, the popular receding-horizon control, model predictive
control (MPC) [17,18], can be used to update a short-term traffic prediction at each time instant,
which effectively approximates the long-term traffic prediction of DP and PMP by means of real-time
trajectory corrections using the current state estimate.
Unlike the internal-combustion engine vehicle where the pulse-and-glide driving strategy is
assumed to the optimal way to increase the fuel economy, it is well-known that speed smoothing of
the BEV can optimize the energy efficiency of the electric motor [19]. Therefore, the speed trajectory
of the BEV should be flatten as much as possible for the prediction horizon to prevent battery spikes,
while satisfying the safety constraints. Also, the battery temperature that significantly influences the
battery efficiency should be maintained in the appropriate ranges. Taking these all into consideration,
we can conclude that the MPC that can handle the state and control constraints is appropriate method
for the battery energy management of the autonomous BEV.
In [20–23], MPC is designed using short-term prediction and it has been confirmed that, to a
certain degree, this method is effective to improve energy efficiency. However, the computational
burdens depending on the several important factors in MPC such as sampling-time and length of
the prediction time were not analyzed thoroughly. To the best of our knowledge, computational load
reduction in MPC for real-time computational feasibility has not been discussed in the literature of
self-driving cars, and this paper will reveal the critical factors determining the computational burden
in MPC. Finally, several remedies to increase the feasibility of the MPC will be suggested.

1.2. Research Contribution

The primary contributions of this paper can be summarized as follows.
First, to overcome the technical shortcomings of a nominal MPC, several methods that enable
the real-time implementation of the optimizer are proposed, and its effectiveness is verified through
various case studies. Depending on the driving conditions, the computational-time constraint violation
rate is reduced around 35% compared to the one of a nominal MPC, but the control performance is
similar to each other, which is our central argument.
Secondly, an appropriate prediction-horizon selection for MPC for BEVs control is studied by
comparing the performances with the one of DP. In most works employing MPC for trajectory
Electronics 2020, 9, 1277 4 of 19

optimization and control, the procedure determining the horizon length is omitted, but our approach
starts with this important aspect considering the implementation in real hardware. As compared to the
control results of nominal MPC and DP, the proposed method can reduce the computational burden
significantly while minimizing the performance degradation in terms of the battery state-of-charge
(SOC) reduction.

1.3. Paper Organization

This paper is organized as follows. Section 2 introduces the vehicle- and battery-dynamics models,
and the optimization control problem is formulated in Section 3. Next, the central part of this paper
is described in Section 4, which proposes a method to reduce the computational complexity of MPC.
Section 6 demonstrates the effectiveness of our approach through simulations, and Section 7 concludes
the paper.

2. Vehicle and Battery Dynamics

Since the interest of this study is to optimize the energy efficiency of the BEVs, relevant vehicle
and powertrain dynamics are modeled first. Note that, compared to other types of vehicles, BEVs
have a simple structure, as shown in Figure 2. Since the power is transmitted through a single path,
the energy loss is only determined by the efficiency of each actuator depending on the operating points.
Before designing the controller, the central parts of BEVs are simplified and modeled in this section.

Figure 2. Schematic of a battery-electric vehicle.

2.1. Vehicle Longitudinal Dynamics

Consider the point-mass longitudinal dynamics of the vehicle. We assume that the effective
mass of the vehicle, which normally accounts for the vehicle static and rotational effect together,
is approximately equivalent to the vehicle mass [24]. Accordingly, the continuous-time longitudinal
dynamics are given by

ṡ = v, (1)
ρAf Cd v2
 
Tm + Tb
v̇ = i0 − − g sin θ − γg cos θ, (2)
mr 2m

where s ∈ R is the vehicle position, v ∈ R is the vehicle speed, m > 0 is the vehicle mass, Tm ∈ R and
Tb ∈ R are the motor and friction brake torques, respectively, r > 0 is the wheel radius, i0 > 0 is the final
gear ratio, ρ > 0 is the air density, Af > 0 is the frontal area of the vehicle, Cd ≥ 0 is the aerodynamic drag
coefficient, θ ∈ (−π/2, π/2) is the road inclination that influences the energy consumption significantly
but assumed to be zero in this paper, g > 0 is the gravitational constant, and γ ≥ 0 is the coefficient of
rolling resistance. Note that θ is a function of the position, that is, θ (s), and that the rolling friction is
kinetic, that is, γ = 0 for v = 0. In the remainder of this paper, we consider Tb = 0 without a significant
loss of generality.
Electronics 2020, 9, 1277 5 of 19

In actual fact, several parameters that are assumed to be constant such as m and θ influence the
vehicle’s energy efficiency significantly. Also, the other constant parameters vary depending on driving
conditions, so the parameter estimations are needed for the accurate energy-efficiency calculation.
However, for simplicity, this paper does not consider the varying-parameters, but it should be treated
in the future.
To apply digital control methods, the continuous-time mode (Equations (1) and (2)) is discretized
using the Euler-forward method
xkv+1 = xkv + gv ( xkv , uvk ) Ts , (3)

where xkv , [sk vk ]T , uvk , Tm,k , gv : R2 × R → R2 is the vehicle dynamics function vector that
aggregates in a column the right-hand side of Equations (1) and (2), and Ts > 0 is the sampling time.

2.2. Battery Dynamics

2.2.1. Continuous-Time SOC Dynamics

The continuous-time battery SOC dynamics is given by

˙ = − Ib ,
SOC (4)
Cb

where SOC ∈ [0, 1] is the state of charge of the battery, Ib ∈ R is the battery current, and Cb > 0 is the
battery capacity and is assumed to be constant.
The battery current Ib is a function of the open-circuit voltage Voc ∈ R, the battery resistance
Rb > 0, and the battery-power consumption Pb ∈ R as,
p
Voc − 2 − 4R P
Voc b b
Ib = , (5)
2 Rb

where 
 Pm
+ for Pm ≥ 0,
ηb
Pb = (6)
 Pm for Pm < 0,
ηb−

ηb+ ∈ (0, 1) is the battery-depletion efficiency, ηb− > 1 is the battery-recharge efficiency, and Pm is the
motor-output power.
The SOC dynamics is captured more accurately by accounting for the variations of Voc and Rb ,
which are functions of SOC, as illustrated in Figure 3.

13.5 0.25

0.2
13

0.15
12.5
0.1

12
0.05

11.5 0
0 0.2 0.4 0.6 0.8 1

Figure 3. Variation of the open-circuit voltage Voc and internal resistance Rb as functions of
state-of-charge (SOC).
Note that, in practice, the variations of Voc and Rb are relatively small compared to Pb during
driving time. Therefore, the battery-output power Pb mainly contributes to the battery SOC changes
Electronics 2020, 9, 1277 6 of 19

in Equation (4). In general, Pb is used as a control input in the battery dynamics. Note that Pb depends
also on the motor-output power Pm , whose dynamics are described hereafter.

2.2.2. Motor Dynamics

The motor-output power is determined by the motor torque Tm , motor speed ωm , vr i0 , and motor
efficiency ηm as
Tm ωm
Pm = . (7)
ηm ( Tm , ωm )
Note that the motor efficiency ηm is a function of the motor operating points, which depend on
Tm and wm (see Figure 4). Furthermore, we assume that the motor efficiency is mirrored in terms of
propulsion and regenerative braking, that is, ηm ( Tm , ωm ) = ηm (− Tm , ωm ).

200 0.7
0.75 0.7
0.75
0.7 0.850.8 0.850.8
0.75
0.
56

0.8
150 0.85 0.
85

100 0.9
Motor Torque [Nm]

0.9
0.7

50
0.9
0.85
0.75

0.8 0.85 0.8

0.8

0 0.8 0.75 0.7 0.7

0.75
0.75 0.7 0.7
0.75
0.8 0.850.8
0.85 0.9
5
0.8
-50 0.9

-100 0.9
0.7

0.85
-150 0.75 0.8
0.
85
0.65

0.85
0.750.8 0.750.8
-200 0.7 0.7

0 2000 4000 6000 8000 10000

Motor Speed [RPM]
Figure 4. Motor efficiency map.

2.2.3. Discrete-Time Battery Dynamics

By combining Equations (4)–(7) and using the Euler-forward method, the continuous-time SOC
dynamics Equation (4) can be discretized as

SOCk+1 = SOCk + gs (SOCk , Tm,k ) Ts , (8)

where gs : [0, 1] × R → [0, 1] is the SOC dynamics function vector, which aggregates as a column the
right-hand side of Equations (4) and (5).

2.3. Discrete-Time Model for Implementation

The discrete-time model of the vehicle and battery dynamics for digital implementation are given by

xk+1 = xk + f ( xk , uk ) Ts , (9)

where xk , [sk vk SOCk ]T , uk , Tm , and f , [( gv )T ( gs )T ]T . In this paper, we assume that, for all
k ≥ 0, the state xk is available, and thus, an observer needs not to be built. Based on Equation (9),
the optimization control problem and several optimization methods will be described in Section 3.
Electronics 2020, 9, 1277 7 of 19

3. Optimal Control Problem Formulation

In this paper, we assume that the controlled car (ego car) is driving between a leader and rear
car. Additionally, the ego car is able to send and receive information through V2X communication
(see Figure 1). In this case scenario, the ego car can plan its future trip by predicting the traffic
dynamics while ensuring safety constraints between the leader and rear cars. Assuming that the
rear car behaves rationally, it is possible to ignore its behavior in the optimization formulation [25].
In addition, this paper assumes only a single lane, which means that no lane changing is considered.
However, in reality, such interactions between the ego vehicle and adjacent vehicles might occur
frequently and influence on the ego vehicle’s trajectory planning. In this paper, however, we exclude
the above-mentioned situations, which are left to our future work.
The objective of the optimization formulation is to minimize the battery-output power over a
given route by optimizing the speed profile.
Let N > 0 be the discrete-time prediction horizon, define x̂i|k , [ŝi|k v̂i|k SOC ˆ i|k ]T as the
i-step-ahead state predicted at step k, let ω̂m,i|k be the i-step-ahead motor speed predicted at step k,
and define Ûk , [û0|k · · · û N −1|k ]T ∈ R N as the sequence of control inputs computed at step k.
Using Equation (9), the optimization problem at step k is given by

N −1
min Jk (Ûk ) =
Ûk
∑ Pb,i (ûi|k , x̂i|k ), (10a)
i =0
s.t.
x̂i+1|k = x̂i|k + f ( x̂i|k , ûi|k ) Ts , (10b)
T
x̂0|k = [sk vk SOCk ] , (10c)
vmin ≤ v̂i+1|k ≤ vmax , (10d)
τmin (v̂i+1|k + δ) ≤ sil+1 − ŝi+1|k ≤ τmax (v̂i+1|k + δ) (10e)
ˆ i+1|k ≤ SOCmax
SOCmin ≤ SOC (10f)
min max
Tm,i (ω̂m,i|k ) ≤ ûi|k ≤ Tm,i (ω̂m,i|k ) (10g)
i = 0, · · · , N − 1, (10h)

where sl is the deterministic position of leader car provided by driving cycle, 0 ≤ vmin < vmax ,
0 < τmin < τmax are the minimum/maximum time headways, δ > 0 is the velocity bound,
SOCmin < SOCmax ∈ [0, 1], and Tm,i min : R → R and T max : R → R are given constraint functions
m,i
at step j that depend on ω̂m,i|k .
Note that the problem formulation (10a)–(10h) is a receding-horizon problem formulation if the
problem must be solved for all k ≥ 0. The terms (10d)–(10f) are the state constraints, and (10g) is the
control constraint. The constraint bounds are reasonable values according to the driving conditions on
US roads [26]. In particular, depending on the ego car’s speed, (10e) indicates the reasonable varying
position constraint between the ego car and leader car.

4. Practical Considerations for Real-Time Implementation of MPC

The receding-horizon control problem (10a)–(10h) can be solved using standard nonlinear solvers.
We expect superior control performance in MPC for long prediction horizons in the case where
the model matches the plant and the solver gives the global optimal solution at each step.
However, for implementation reasons, an appropriate prediction-horizon length should be chosen.
In this section, we first give the performance benchmark of DP when the prediction horizon is the
entire trip of the ego car. Next, for real-time implementation in MPC, we propose a simplification of
the cost function in order to reduce computational complexity. A numerical analysis is carried out to
highlight the impact of an increase in N on performance and computational feasibility of MPC.
Electronics 2020, 9, 1277 8 of 19

All simulations presented in this paper are carried out using MATLAB R2020a, Windows 10,
Intel Core i7-9700 CPU @ 3.00 Ghz, and 16 GB RAM. Furthermore, the continuous-time dynamics are
integrated by ode45 and the control is constant between samples.

4.1. Dynamic Programming

In this subsection, the deterministic DP algorithm is employed to solve (10a)–(10h) for the entire
trip. In general, DP is not applicable in practice because DP requires all future reference inputs
and disturbances in advance, indicating that DP is a noncausal controller [27,28]. For example,
the entire speed trajectory of the leader car should be provided, which is not reasonable in reality.
However, in many applications, the control performance using DP is usually assumed to be the best
solution that guarantees global optimality. Therefore, employing DP is meaningful as a benchmark,
with which the control performance of other causal controllers can be compared. In this paper, we will
compare MPC with DP benchmark performance.
DP solves the optimization problem (10a)–(10h) by setting k = 0 and N = Ndp , where Ndp > 0 is
the discrete-time prediction horizon of the entire trip. Note that, as DP is used once and thus does not
depend on the actual time step k, we omit the index k in (10a)–(10h) in this subsection.
For all j ≥ 0, let X j be the feasible state-constraint set (10d)–(10f) at predicted step j, let X N be the
terminal state-constraint set, and let U j be the feasible control-constraint set (10g) at predicted step j.
Next, define the discretized-state set X̃ , { x̃1 , . . . , x̃i , . . . , x̃` } ⊂ X j , for all 0 ≤ j ≤ N, where x̃i is the
i-th node of X̃, i ∈ {1, . . . , `}, and ` > 0 is the total number of discretized states in X̃.
Using the principle of optimality [29], the optimal control policy for the node x̃i is obtained by
backward induction, where x̂0 = x̃i in (10c). In particular, starting from the optimal final stage cost

J N ( x̃i ) = min Pb,N (û N , x̃i ), (11)

û N ∈U N

where N = Ndp , the DP backward recursion is given by

J j ( x̃i ) = min [ Pb,j (û j , x̃i )

û j ∈U j
(12)
+J j+1 ( x̃i + f ( x̃i , û j ) Ts )],

where j ∈ { N − 1, . . . , 0}. Accordingly, the optimal control policy π for each node x̃i is mapped by

π = {µ0 ( x ), µ1 ( x ), · · · , µ Ndp −1 ( x )}, (13)

where µi : R3 → R is the mapping function and i = 0, . . . , Ndp − 1. Accordingly, Equation (13) can be
used to find the optimal control input in a forward simulation.
In this paper, we use the generic DP function in Matlab, that is, dpm [30]. An appropriate grid
number for the state and control variables must be specified to avoid the curse of dimensionality.
Doing so, the computation is completed approximately in five minutes while satisfying all constraints,
as illustrated in Figure 5. The entire vehicle speed trajectories for two drive cycles, i.e., WLTC, US06,
are optimized so as to minimize the cost and enforce the constraints. One distinguished feature is
that the speed profile when applying DP is flattened as much as possible to prevent spikes in the
battery current.
The battery SOC consumption for each drive cycle is compared in Table 1, which shows that DP
can significantly improve the energy efficiency of BEV around 14 to 20%. These control performances
are considered ideal, so we will use these as our benchmark. However, as mentioned, DP is not
implementable in real time due to the computational complexity and strong assumptions on command
and disturbance preview.
Electronics 2020, 9, 1277 9 of 19

(a): WLTC Drive Cycle

40

30
Baseline
DP
20

10

0
0 200 400 600 800 1000 1200 1400 1600 1800

(b): US06 Drive Cycle

40

30

20

10

0
0 100 200 300 400 500 600

Figure 5. Comparison of the speed trajectories obtained by dynamic programming (DP) and
human-driver maneuvers with (a) WLTC and (b) US06.

Table 1. Comparison of the battery state-of-charge (SOC) consumption (%) with the speed profiles
obtained by dynamic programming (DP) and nominal speed.

Drive Cycle Baseline DP Improvement (%)

WLTC 17.55 14.96 14.76
US06 13.41 10.74 19.90

Since the cost function is nonconvex and the dynamics are nonlinear, several local minima may
be presented. However, as DP computes all costs for every node by backward induction, it can be said
that the obtained performance ensures global optimality.

4.2. Model Predictive Control

4.2.1. Quadratic Cost Simplification

In the previous section, the nonlinear function Pb is used as a cost when applying DP, which gives a
globally optimal result by backward induction. However, for real-time computation, a computationally
tractable method is required. In particular, we show in this section that the nonlinear cost Pb can be
approximated by a quadratic cost.
In [19], by assuming constant actuator efficiencies, the motor power Pb can be simplified as

Tm v 2
Pb ≈ Pm ≈ i0 + αTm (14)
r
where α > 0 is a tunable motor parameter. Note that, even though the cost is greatly simplified
compared to the original Pb , Equation (14) is still nonlinear and depends on the value of α, which
modifies the overall control performance significantly.
In this paper, the cost function is further simplified as follows. Since the objective of this study
is to minimize the battery SOC reduction, the fluctuation of Ib should be minimized according to
Equation (4). It also can be interpreted that the sudden changes in the control input u, which cause Ib
Electronics 2020, 9, 1277 10 of 19

fluctuations, should be avoided as much as possible. Therefore, for all k ≥ 0, another minimization of
a quadratic cost in place of Equation (10a) is used by eliminating the first term in Equation (14), that is,

N −1 N −1
min ∑
ûi|k ∈Ui|k i =0
Pb,i ( x̂i|k , ûi|k ) ≈ min ∑
ûi|k ∈Ui|k i =0
û2i|k , (15)

where Ui|k is the i-step-ahead constraint Equation (10g) at step k.

To verify the above assumption, the simulation results using the cost functions (10a) and (15) are
compared in Figure 6. It is important to note that the SOC reductions for both cases are almost the
same. Due to the nonconvexity in Pb , the control fluctuates a lot to search the optimum point at every
step, and is sometimes stuck at the local minimum points, as shown in the middle plot of Figure 6.

0.8

0.79

0.78

0.77

0.76
0 20 40 60 80 100 120 140 160 180 200

300

200

100

-100

-200

-300
0 20 40 60 80 100 120 140 160 180 200

15

10

0
0 20 40 60 80 100 120 140 160 180 200

Figure 6. Comparison of the simulation results using the different cost functions with N = 10 and
Ts = 1 s. The top plot shows the battery SOC trajectories, the middle plot shows the control input
evolutions, and the bottom plot shows the computation times tc,k at each step k.

In contrast, by exploiting the quadratic form (15), the computation time at each step is significantly
reduced compared to the original form, as shown in the bottom plot of Figure 6.
Electronics 2020, 9, 1277 11 of 19

Therefore, it can be concluded that the nonlinear cost (10a) can be replaced with the quadratic
cost (15) without sacrificing too much performance in terms of the battery SOC reduction.

4.2.2. Real-Time Computational Feasibility

One of the major issues in MPC is the computational burden when we handle large or fast
dynamics-related systems. Moreover, a long prediction horizon increases the computational complexity
significantly, even though a longer horizon provides better performance in the ideal case.
In this subsection, an appropriate prediction horizon for our problem that takes into account the
relationship between performance and real-time implementation feasibility is analyzed. The nonlinear
solver fmincon in MATLAB is used to solve the optimization control problem (10a)–(10h), where the
sampling time is specified as Ts = 1 s in all cases. That is, the computation should be completed in 1 s
at each step.
By solving the optimization problem, we obtain the optimal sequence Ûk∗ at each step k and
only the first element û0∗|k is applied, which is the principle of receding horizon control. To see the
relationship between the control performance and the prediction horizon, we gradually increase the
prediction horizon while observing the required computations for each horizon. Note that it is possible
to achieve superior performance with faster computation by using other advanced solvers, e.g., tailored
NLP solvers using inexact and real-time iteration solution strategies [31]. However, for the purpose of
comparison, we utilize the same solver for all simulations.
As shown in Figure 7, better control performance in terms of the battery SOC reduction can be
achieved by increasing N. However, the required average computation time for each step exceeds the
sample time Ts , meaning that fmincon is not implementable in real time.

(a)

3.85

3.8

3.75
5 6 7 8 9 10 11 12 13 14 15

(b)
100

50

0
5 6 7 8 9 10 11 12 13 14 15

Figure 7. (a) Battery SOC reductions with respect to N, and (b) computational burdens with respect to
N. The computational-time constraint violation represents the ratio between the number of occurrences
where tc > Ts and the number of samples of the entire experiment.

It can be concluded that N = 10 is sufficient to optimize the energy efficiency of our BEV, but the
computational burden should be reduced for implementation. A collection of methods to deal with
the computational complexity of MPC is proposed in the next section.

5. Strategies for Computational Load Reduction in MPC

In the previous section, an appropriate prediction horizon for MPC is determined heuristically.
The computational load is, however, unmanageable with the specified horizon. In this section,
we suggest a collection of strategies, which potentially reduce the computational load of MPC.
Electronics 2020, 9, 1277 12 of 19

In particular, we describe sampling-time adjustment, warmstarting, and move-blocking strategies,

and explain their possible impact upon performance.

5.1. Sampling-Time Adjustment

A sampling-time increase is expected to reduce the computational load of MPC as, for a similar
preview of the trajectories, the prediction horizon N decreases. However, note that, as Ts increases,
crucial intersampled information might be ignored during prediction causing potential performance
degradation of MPC, especially regarding constraint satisfaction. In the following, we numerically
investigate the impact of larger sampling times in MPC.
As shown in Figure 8, the position constraints between the ego car and leader car, which are
hard constraints, are sometimes violated as Ts increases. As expected, we observe that the average
computation time proportionally decreases as Ts increases, and the result is omitted since it is obvious.
In fact, due to the inaccuracy of intersampling model information, MPC cannot handle the constraints
when the vehicle is close to the bounds (see e.g., at 40 s in Figure 8b,c).

(a)
60

40

20

0
0 50 100 150 200

(b)
60

40

20

0
0 50 100 150 200

(c)
60

40

20

0
0 50 100 150 200

Figure 8. Distance difference between ego car and leader car with (a) Ts = 1 s, (b) Ts = 2 s, and (c)
Ts = 3 s.

5.2. Warmstarting
Another remedy for computational burden reduction is warmstarting [32]. Warmstarting is a
method that supplies an initial guess of the sequence of controls to the solver rather than using a
constant initial control sequence, typically, a vector of zeros. In general, supplying an initial guess is
useful when dealing with nonconvex problems and nonlinear solvers because the solvers can easily
get stuck to a local minimum point. In addition, this scheme also prevents cases with singularities by
avoiding regions close to the initial condition.
Electronics 2020, 9, 1277 13 of 19

Specifically, instead of starting the optimizer with the initial control sequence Ûk,0 = [0 · · · 0]T ,
we use the previous optimal solution, which is computed at k − 1, at the current step k as
(
[0 · · · 0]T , k = 0,
Ûk,0 = (16)
[Ûk∗−1 (2 : end) Ûk∗−1 (end)]T , k > 0,

where Ûk∗−1 (2 : end) , [u1∗|k−1 · · · u∗N −1|k−1 ]T ∈ R N −1 is the optimal control sequence computed at
step k − 1, which is truncated from the second component to the N-th component. Note that the last
element Ûk∗−1 (end) is copied twice in Ûk,0 so as to match the number of components in the array to
N components.
The effectiveness of adopting warmstarting strategy is summarized in Table 2. Although the
improvement is not very significant, it can be said that adopting warmstarting is usually helpful.

Table 2. Comparison of the average computation time with and without warmstarting.

Without Warmstarting With Warmstarting

N = 15 2.5336 s 2.5210 s
N = 20 2.5646 s 2.5512 s

5.3. Move Blocking

To reduce the computational load of MPC, a move-blocking strategy is considered [33].
Move-blocking strategy is similar to control-horizon strategy, where the control is free to move
for the first initial prediction steps and then blocked. The difference is that, after the free initial moves
of the control, move-blocking strategy blocks the control for several consecutive intervals, whereas
control-horizon strategy blocks the last control for the remainder of the prediction horizon.
Control-horizon strategy is described as follows. At step k, instead of solving the optimization
problem with the optimal control sequence Ûk ∈ R N , we use the alternative sequence Ûk0 ∈ R Nc +1 ,
where Nc is the control horizon and 0 < Nc < N. The relationship between Ûk and Ûk0 is given by

Ûk = Tc Ûk0 , (17)

where Tc ∈ R N ×( Nc +1) is the control-horizon matrix and is defined by

Tc , blkdiag( INc , 1( N − Nc )×1 ) ∈ R N ×( Nc +1) , (18)

where In is the n-by-n identity matrix, and 1m×n is the m-by-n matrix of ones.
A variation of the control-horizon strategy is adopted in this paper, which is the
move-blocking strategy. The motivation of this variation is that, since the reference position of
the leader car can vary significantly in the future, blocking the control over a large interval yields
constraint violation and, in some cases, suboptimality or infeasibility. Therefore, we divide the
prediction horizon into several blocking intervals. Accordingly, the control-horizon matrix is modified
to the move-blocking matrix

Tb = blkdiag( Ikb , 1kb ×1 , . . . , 1kb ×1 , 1kr ×1 ) ∈ R N × Nb , (19)

where 0 < kb < N is the number of control moves blocked at each interval (except the first interval), kr
is the remainder of the division N/kb (if zero, then this term disappears), and Nb , ceil( N/kb ) − 1 + kb
is the length of Ûk0 using move-blocking strategy. With this move-blocking matrix, the control is free to
move for the first kb steps, and, afterwards, blocked for kb steps consecutively within Nb − kb intervals.
An example of move-blocking sequence using the proposed method is shown in Figure 9 with kb = 3
and N = 10.
Electronics 2020, 9, 1277 14 of 19

120

100

80

60

40

20

0
0 1 2 3 4 5 6 7 8 9

Figure 9. Control sequence using move-blocking strategy, where kb = 3 and N = 10. The first three
moves are free, then blocked in consecutive blocks of three controls. Note that Nb = 6, kr = 1, and the
last control is within a separate block from k = 9 to k = 10.

6. Simulation Results
So far, several methods to reduce the computational burden of MPC have been discussed. In this
section, the effectiveness of the proposed strategies is verified for two driving cycles, namely, WLTC
and US06. The parameters of the simulations are listed in Table 3. The model parameters are extracted
from the Ford Focus BEV model in ADVISOR [34], which is a high-fidelity simulator with the aim of
analyzing the vehicle-energy efficiency.

Table 3. Simulation Parameters.

Symbol Description Value (Unit)

m Vehicle total mass 1445 (kg)
r Wheel radius 0.3166 (m)
Af Vehicle frontal area 2.06 (m2 )
Cd Aerodynamic drag coefficient 0.312
ρ Air density 1.2 (kg/m3 )
θ road inclination 0 (◦ )
γ Rolling resistance coefficient 0.0086
i0 Final drive ratio 4.2
vmin , vmax Acceptable range of speed (0, 150} (km/h)
Cb Battery capacity 55 (Ah)
ηb+ Battery-depletion efficiency 0.9
ηb− Battery-recharge efficiency 1.11
τmin , τmax max/min time headway {1, 2} (s)
N Prediction horizon 10
Ts Sampling time 1 (s)
kb Control moves blocked 3

The following simulations compare control performance between the proposed method and
other strategies, i.e., nominal MPC and DP. As mentioned earlier, the performance using DP can be
considered as ideal. Therefore, if the performance using our approach closely matches the one of DP
with a computationally manageable complexity, it can be said that the main objective of this study is
achieved. Also, compared to the nominal MPC, the computational complexity should be reduced with
the almost same control performance.
Simulation 1: WLTC driving cycle. The results in urban-driving scenario obtained from the proposed
method and DP are compared in Figure 10. Since the state and control trajectories using the nominal
MPC are very similar to those of our approach, we omit the plots with the nominal MPC.
Electronics 2020, 9, 1277 15 of 19

(a)
100

80

60

40

20

0
0 200 400 600 800 1000 1200 1400 1600 1800

(b)
40

30

20

10

0
0 200 400 600 800 1000 1200 1400 1600 1800

(c)
0.8

0.75

0.7

0.65

0.6
0 200 400 600 800 1000 1200 1400 1600 1800

(d)
300

200

100

-100

-200

-300
0 200 400 600 800 1000 1200 1400 1600 1800

Figure 10. Comparison between model prediction control (MPC) and DP for the WLTC driving cycle,
(a) distance difference between ego car and leader car, (b) speed profiles, (c) battery SOC trajectories,
and (d) control trajectories.

Note that the appropriate distances between the self-driving car and leader car are maintained
for both methods, as shown in Figure 10a. Depending on the self-driving car speed, the time-varying
upper- and lower- bounds (10e) are specified in the predictions. No constraint violation occurs for MPC
and DP. The entire speed trajectories are very similar to each other, but slightly different in Figure 10b.
The battery SOC trajectories are depicted in Figure 10c and the SOC consumption using the proposed
method is very similar to that of DP, which verifies the effectiveness of our approach. Since DP
explores all possible combinations of control and state pairs, the control input in Figure 10d sometimes
changes rapidly because DP is just designed to minimize the cost function without considering the
physical relationship between the actuator and vehicle. Therefore, the control performance using DP is
exaggerated to some extent, but it can serve as a benchmark that only considers the minimization of
the cost.
The trajectories of s, v, and SOC are shown in Figure 11a–c. From the battery SOC trajectory
in Figure 11c, the proposed MPC, which is computationally tractable, can give a solution that is
very close to the globally optimum solution obtained by DP. Moreover, the trajectories obtained by
DP sometimes physically do not make sense and should be corrected for hardware implementation.
Since the nonlinear model used in this paper only captures the primary dynamics of the vehicle and
powertrain, the control input using DP in Figure 11d cannot ensure a globally optimum solution in
Electronics 2020, 9, 1277 16 of 19

real production vehicle. Therefore, in reality, we expect that the performance gap between DP and our
approach is much closer to each other, but the computational burden is significantly reduced with our
approach, which is the central argument of this paper.
Simulation 2: US06 drive cycle. The US06 drive cycle represents the case of a highway-driving scenario.
The results are shown in Figure 11.

(a)
80

60

40

20

0
0 100 200 300 400 500 600

(b)
40

30

20

10

0
0 100 200 300 400 500 600

(c)
0.8

0.78

0.76

0.74

0.72

0.7

0 100 200 300 400 500 600

(d)
300

200

100

-100

-200

-300
0 100 200 300 400 500 600

Figure 11. Comparison between MPC and DP in US06 driving cycle, (a) distance difference between
ego-car and leader-car, (b) speed profiles, (c) battery SOC trajectories, and (d) control trajectories.

Discussion. Table 4 and Figure 12 support the above claims. As expected, the speed modification that
minimizes Tm 2 is effective in terms of battery SOC reduction. As expected, the proposed MPC with

move-blocking method does not outperform the nominal MPC but is very close to that of nominal MPC
in terms of battery SOC reduction performance (see Table 4). Unfortunately, the control performances
of our approach and nominal MPC shows the significant difference by comparing with DP result,
but DP is not implementable in the real-time and trajectories from the DP do not consider the physical
limitation of the actuator.
Electronics 2020, 9, 1277 17 of 19

Table 4. ∆SOC obtained by different optimized speed trajectories (the values in parentheses describe
the improvements compared to the baseline).

WLTC US06
Baseline 17.55 13.41
Proposed 15.64 (10.88%) 11.42 (14.83%)
Nominal MPC 15.42 (12.14%) 11.30 (15.73%)
DP 14.96 (14.76%) 10.74 (19.90%)

Nominal MPC
40
Proposed

30

20

10

0
US06 WLTC

Figure 12. Comparison of the computational-time constraint violation rates obtained by nominal MPC
and proposed method.

Note that the computational-time constraint violation rate is significantly reduced with the
proposed method (see Figure 12). Even though our approach sometimes exceeds the sampling
period Ts , some remedies to handle this issue are possible. For example, we can further reduce the
computation time by replacing the language from MATLAB to C, which is essentially needed for
hardware implementation. Also, constraint relaxation and the introduction of faster solvers instead of
fmincon can be alternative solutions, which are left to our future works.

7. Conclusions
There have been extensive approaches in the literature to apply MPC in automotive control
applications, but its computational feasibility was not clearly reviewed. This paper investigated several
methods to reduce the computational complexity of a nominal MPC implemented in a battery-electric
vehicle (BEV). First, an appropriate prediction horizon for our optimization control problem was
determined with a nominal MPC, and we concluded that the control performance of nominal MPC
closely matches the one of DP, which is assumed to be globally optimal. Next, to ensure real-time
feasibility, adopting move-blocking strategy, with warmstarting and large sampling times, significantly
reduces the computational burden compared to a nominal MPC without sacrificing too much of the
control performance in terms of battery state-of-charge (SOC) reduction. Methods that provide further
reduction of the computation load for real-time implementation of MPC in hardware will be our future
work. In addition, a lot of constraint violations are expected to occur with the proposed approach
when the reference value varies significantly. To overcome such a shortcoming, the relaxation of the
constraints with the proposed approach is also left to our future work.

Author Contributions: Conceptualization, K.H., T.W.N., and K.N.; Data curation, K.H. and T.W.N.; Formal
analysis, K.H.; Funding acquisition, K.N; Investigation, K.H. and T.W.N.; Methodology, K.H., and T.W.N.;
Software, K.H. and T.W.N.; Supervision, K.N.; Visualization, T.W.N.; Writing—original draft, K.H. and T.W.N.;
Writing—review & editing, K.N. All authors have read and agreed to the published version of the manuscript.
Electronics 2020, 9, 1277 18 of 19

Funding: This paper was supported by Korea Institute for Advancement of Technology(KIAT) grant funded by
the Korea Government(MOTIE)(P0008473, HRD Program for Industrial Innovation).
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Wadud, Z.; MacKenzie, D.; Leiby, P. Help or hindrance? The travel, energy and carbon impacts of highly
automated vehicles. Transp. Res. Part A Policy Pract. 2016, 86, 1–18. [CrossRef]
2. Masood, K.; Molfino, R.; Zoppi, M. Simulated Sensor Based Strategies for Obstacle Avoidance Using Velocity
Profiling for Autonomous Vehicle FURBOT. Electronics 2020, 9, 883. [CrossRef]
3. Li, N.; Oyler, D.W.; Zhang, M.; Yildiz, Y.; Kolmanovsky, I.; Girard, A.R. Game theoretic modeling of driver
and vehicle interactions for verification and validation of autonomous vehicle control systems. IEEE Trans.
Control Syst. Technol. 2017, 26, 1782–1797. [CrossRef]
4. Han, K.; Li, N.; Kolmanovsky, I.; Girard, A.; Wang, Y.; Filev, D.; Dai, E. Hierarchical Optimization of Speed
and Gearshift Control for Battery Electric Vehicles Using Preview Information. In Proceedings of the IEEE
American Control Conference (ACC), Denver, CO, USA, 1–3 July 2020; pp. 4913–4919.
5. Han, J.; Sciarretta, A.; Ojeda, L.L.; De Nunzio, G.; Thibault, L. Safe-and eco-driving control for connected
and automated electric vehicles using analytical state-constrained optimal solution. IEEE Trans. Intell. Veh.
2018, 3, 163–172. [CrossRef]
6. Ersal, T.; Kolmanovsky, I.; Masoud, N.; Ozay, N.; Scruggs, J.; Vasudevan, R.; Orosz, G. Connected and
automated road vehicles: State of the art and future challenges. Veh. Syst. Dyn. 2020, 58, 672–704. [CrossRef]
7. Tate, L.; Hochgreb, S.; Hall, J.; Bassett, M. Energy Efficiency of Autonomous Car Powertrain; Technical Report;
SAE Technical Paper; SAE International: Warrendale, PA, USA, 2018.
8. Chen, T.D.; Kockelman, K.M.; Hanna, J.P. Operations of a shared, autonomous, electric vehicle fleet:
Implications of vehicle & charging infrastructure decisions. Transp. Res. Part A Policy Pract. 2016, 94, 243–254.
9. Zeng, X.; Wang, J. Globally energy-optimal speed planning for road vehicles on a given route. Transp. Res.
Part C Emerg. Technol. 2018, 93, 148–160. [CrossRef]
10. Chen, D.; Kim, Y.; Stefanopoulou, A.G. State of charge node planning with segmented traffic information.
In Proceedings of the IEEE Annual American Control Conference (ACC), Milwaukee, WI, USA, 27–29 June 2018;
pp. 4969–4974.
11. Wan, N.; Vahidi, A.; Luckow, A. Optimal speed advisory for connected vehicles in arterial roads and the
impact on mixed traffic. Transp. Res. Part C Emerg. Technol. 2016, 69, 548–563. [CrossRef]
12. McDonough, K.; Kolmanovsky, I.; Filev, D.; Szwabowski, S.; Yanakiev, D.; Michelini, J. Stochastic fuel
efficient optimal control of vehicle speed. In Optimization and Optimal Control in Automotive Systems; Springer:
Berlin, Germany, 2014; pp. 147–162.
13. Mehta, P.; Meyn, S. Q-learning and Pontryagin’s minimum principle. In Proceedings of the 48h IEEE
Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference, Shanghai,
China, 15–19 December 2009; pp. 3598–3605.
14. Lee, H.; Song, C.; Kim, N.; Cha, S.W. Comparative Analysis of Energy Management Strategies for HEV:
Dynamic Programming and Reinforcement Learning. IEEE Access 2020, 8, 67112–67123. [CrossRef]
15. Lee, H.; Kang, C.; Park, Y.I.; Kim, N.; Cha, S.W. Online Data-Driven Energy Management of a Hybrid Electric
Vehicle Using Model-Based Q-Learning. IEEE Access 2020, 8, 84444–84454. [CrossRef]
16. Li, S.; Li, N.; Girard, A.; Kolmanovsky, I. Decision making in dynamic and interactive environments based
on cognitive hierarchy theory: Formulation, solution, and application to autonomous driving. arXiv 2019,
arXiv:1908.04005.
17. Kouvaritakis, B.; Cannon, M. Model Predictive Control; Springer: Charm, Switzerland, 2016.
18. Nguyen, T.W.; Islam, S.A.U.; Bruce, A.L.; Goel, A.; Bernstein, D.S.; Kolmanovsky, I.V. Output-Feedback
RLS-Based Model Predictive Control. In Proceedings of the American Control Conference (ACC), Denver,
CO, USA, 1–3 July 2020.
19. Han, J.; Vahidi, A.; Sciarretta, A. Fundamentals of energy efficient driving for combustion engine and electric
vehicles: An optimal control perspective. Automatica 2019, 103, 558–572. [CrossRef]
Electronics 2020, 9, 1277 19 of 19

20. Prakash, N.; Cimini, G.; Stefanopoulou, A.G.; Brusstar, M.J. Assessing fuel economy from automated driving:
Influence of preview and velocity constraints. In Proceedings of the ASME 2016 Dynamic Systems and
Control Conference, Boston, MA, USA, 6–8 July 2016.
21. Seok, J.; Wang, Y.; Filev, D.; Kolmanovsky, I.; Girard, A. Energy-Efficient Control Approach for Automated
HEV and BEV With Short-Horizon Preview Information. In Proceedings of the ASME 2018 Dynamic Systems
and Control Conference, Atlanta, GA, USA, 30 September–3 October 2018.
22. HomChaudhuri, B.; Vahidi, A.; Pisu, P. Fast model predictive control-based fuel efficient control strategy for
a group of connected vehicles in urban road conditions. IEEE Trans. Control Syst. Technol. 2016, 25, 760–767.
[CrossRef]
23. Jia, Y.; Jibrin, R.; Itoh, Y.; Görges, D. Energy-optimal adaptive cruise control for electric vehicles in both time
and space domain based on model predictive control. IFAC-PapersOnLine 2019, 52, 13–20. [CrossRef]
24. Kim, Y.; Figueroa-Santos, M.; Prakash, N.; Baek, S.; Siegel, J.B.; Rizzo, D.M. Co-optimization of speed
trajectory and power management for a fuel-cell/battery electric vehicle. Appl. Energy 2020, 260, 114254.
[CrossRef]
25. Brackstone, M.; McDonald, M. Car-following: A historical review. Transp. Res. Part F Traffic Psychol. Behav.
1999, 2, 181–196. [CrossRef]
26. National Highway Traffic Safety Administration. Summary of State Speed Laws Ninth Edition: Current as of
January 1, 2006; National Committee on Uniform Traffic Laws and Ordinances: Washington, DC, USA, 2006.
27. Angel, E. Dynamic programming for noncausal problems. IEEE Trans. Autom. Control 1981, 26, 1041–1047.
[CrossRef]
28. Kalia, A.V.; Fabien, B.C. On Implementing Optimal Energy Management for EREV using Distance Constrained
Adaptive Real-Time Dynamic Programming. Electronics 2020, 9, 228. [CrossRef]
29. Bellman, R. Dynamic programming. Science 1966, 153, 34–37. [CrossRef]
30. Sundstrom, O.; Guzzella, L. A generic dynamic programming Matlab function. In Proceedings of the IEEE
Control Applications, (CCA) & Intelligent Control, (ISIC), St. Petersburg, Russia, 8–10 July 2009; pp. 1625–1630.
31. Walker, K.; Samadi, B.; Huang, M.; Gerhard, J.; Butts, K.; Kolmanovsky, I. Design Environment for Nonlinear
Model Predictive Control; Technical Report; SAE Technical Paper; SAE International: Warrendale, PA, USA, 2016.
32. Liao-McPherson, D.; Huang, M.; Kolmanovsky, I. A regularized and smoothed fischer–burmeister method
for quadratic programming with applications to model predictive control. IEEE Trans. Autom. Control
2018, 64, 2937–2944. [CrossRef]
33. Cagienard, R.; Grieder, P.; Kerrigan, E.C.; Morari, M. Move blocking strategies in receding horizon control.
J. Process Control 2007, 17, 563–570. [CrossRef]
34. Wipke, K.B.; Cuddy, M.R.; Burch, S.D. ADVISOR 2.1: A user-friendly advanced powertrain simulation using
a combined backward/forward approach. IEEE Trans. Veh. Technol. 1999, 48, 1751–1761. [CrossRef]

c 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).