A New Energetically Optimized Power Supply System for a Mobile

Robot Platform, using Ultracapacitors and Batteries to Ensure Both

Ultra-fast Charging and Autonomy

Carlos Arantes

, João Sena Esteves

and João Sepúlveda

Department of Industrial Electronics, University of Minho, Campus of Azurém, 4800-058 Guimarães, Portugal

Centre Algoritmi, Department of Industrial Electronics, University of Minho, Campus of Azurém,

4800-058 Guimarães, Portugal

Keywords: Ultracapacitors, Batteries, Fast Electric Charger, Mobile Robot Platform, Energy Management.

Abstract: The smallest charging times required by fully discharged conventional batteries are some tens of minutes.

This is an important limitation for mobile robot platforms. A previous paper already validated the possibility

of integrating ultracapacitors and batteries in the same system. However, it has some significant limitations:

1) It works with an ultracapacitors module or a battery, but it does not work with both devices at the same

time; 2) It requires an external dedicated charging station; 3) It is not possible to take profit from a part –

which is non-negligible – of the energy previously stored in the ultracapacitors. This paper presents a new

power supply system for mobile robot platforms that has been developed in order to overcome these

limitations. Its main goals are evaluating the feasibility of: 1) Fully integrating batteries and ultracapacitors,

working simultaneously as energy-storing devices, with the aim of enabling a mobile robot platform to

achieve a reasonable autonomy after a very reduced charging time and considerable autonomy when there

are no charging time constraints; 2) Installing all the system in the mobile robot platform, avoiding the use

of an external dedicated charging station; 3) Extracting almost all the energy previously stored in the

ultracapacitors. Both simulation results and experimental results are presented.

1 INTRODUCTION

Even with fast charging techniques, the smallest

charging times required by fully discharged

conventional batteries are some tens of minutes, due

to the allowable values of their charging currents.

An ultra-fast nickel-cadmium battery charger

presented by Petchjatuporn et al. (2005) took about

13 minutes to fully charge a battery at a C-rate of

8C. This C-rate is so high that it considerably

reduces the lifespan of the battery. A research on

fast charging for lead-acid batteries was conducted

by Siguang et al. (2009): charging starts at a C-rate

of 1C and then goes decreasing as the battery is

being charged, taking at least a couple of hours to

fully charge the battery. Batteries charging times of

some tens of minutes are an important limitation for

mobile robot platforms. It is not possible to work

around this limitation by increasing charging

currents, since this procedure would rise the

temperature too much, deteriorating the batteries.

An interesting possibility would be replacing

batteries by ultracapacitors, which allow charging

and discharging currents substantially higher than

those allowed by common batteries. By that way,

their charging periods may be much smaller than

those required by batteries and they are able to

provide power peaks of significant value. However,

using ultracapacitors as the only energy-storing

device in mobile robot platforms is not feasible, due

to the severe limited achievable autonomy

(Muffoletto et al., 2010).

Several applications integrating batteries and

ultracapacitors on the same system are currently

available (Schneuwly and Gallay, 2000; Awerbuch

and Sullivan, 2010; Haifeng and Xueyu, 2010;

Niemoeller and Krein, 2010; Monteiro et al., 2011;

Qin and Zhu, 2011; Musat et al., 2012). This

integration results from the need of having a single

power supply system capable of, simultaneously: 1)

Storing large amounts of energy in the batteries for

the sake of autonomy; 2) Providing power peaks of

short duration but significant value using the

ultracapacitors, extending the battery lifespan

155

Arantes C., Sena Esteves J. and Sepúlveda J..

A New Energetically Optimized Power Supply System for a Mobile Robot Platform, using Ultracapacitors and Batteries to Ensure Both Ultra-fast

Charging and Autonomy.

DOI: 10.5220/0005531201550163

In Proceedings of the 12th International Conference on Informatics in Control, Automation and Robotics (ICINCO-2015), pages 155-163

ISBN: 978-989-758-122-9

 2015 SCITEPRESS (Science and Technology Publications, Lda.)

(Musat et al., 2012) and/or ensuring the proper

functioning of the system (Schneuwly and Gallay,

2000; Monteiro et al., 2011).

A previous paper by Arantes et al., (2014) also

validates the possibility of integrating ultracapacitors

and batteries in the same system. The main goal of

the suggested integration is enabling a mobile robot

platform to achieve 1) a reasonable autonomy after a

very reduced charging time, when its energy-storing

devices (batteries and ultracapacitors) become

discharged and there is not enough time to properly

charge the batteries; 2) a considerable autonomy

when there are no charging time constraints. The

power supply system presented in that paper

achieved satisfactory results but it has some

important limitations, namely:

 It works with an ultracapacitors module or a

battery, but it does not work with both devices at

the same time. It must be shut down in order to

switch the power sources;

 It requires an external dedicated charging station.

In fact, some of its components are external to

the mobile robot platform, for example a heavy

1500VA/50Hz power transformer;

 It is impossible to extract the energy stored in the

ultracapacitors module whenever its voltage

becomes under 1.5V. Therefore, it is not possible

to take profit from a part of the energy

previously stored in that device. In some

applications – for example, the one described by

Kularatna and Patel (2014) – the ultracapacitors

non-extractable energy may be non-negligible.

This paper presents a new power supply system

that has been developed in order to overcome these

limitations. Its main goals are evaluating the

feasibility of:

 Fully integrating batteries and ultracapacitors,

working simultaneously as energy-storing

devices, to enable a mobile robot platform to

achieve a reasonable autonomy after a very

reduced charging time and large autonomy when

there are no charging time constraints;

 Installing all the system in the mobile robot

platform, avoiding the use of an external

dedicated charging station;

 Extracting almost all the energy previously

stored in the ultracapacitors.

The new power supply system is described in

Section 2. The simulation results presented in

Section 3 were performed before implementing the

physical system described in Section 4. The

experimental results presented in Section 5 were

obtained with the real system implementation.

Section 6 provides the general conclusions and some

considerations regarding future developments.

2 SYSTEM ARCHITECTURE

OVERVIEW

The new power supply system has four main

components:

 The energy-storing devices, which are an

ultracapacitors module (116F, 16V, 4.12Wh) and

a lead-acid battery (12V, 12Ah, 144Wh);

 The charger, used to transfer energy from the

electrical grid to the energy-storing devices;

 The energetically optimized voltage regulator,

whose function is to manage the available energy

on the energy-storing devices, ensuring 12.5V on

its output;

 The output voltage regulator, used to supply a

regulated voltage to each part of the mobile robot

platform from de 12.5V established by the

energetically optimized voltage regulator.

2.1 The Charger

Figure 1 shows the block diagram of the subsystem

that uses the charger. It is able to charge the

ultracapacitors module and the lead-acid battery, one

at a time. Firstly, when the charger is turned on (by

connecting it to the electrical grid), it injects current

on the ultracapacitors module as quickly as possible

(the maximum current value is 14A and the

maximum output power value is 80W) in order to

ensure ultra-fast charging. The ultracapacitors

charge process is considered completed when the

voltage goes up to 15.5V. Secondly, when the

ultracapacitors module is charged, the charger

injects 1.2A on the lead-acid battery while the

voltage does not go up to 13.8V. Thirdly, when the

battery voltage achieves 13.8V, the charger keeps

that voltage level constant until the current goes

down to 50mA. Finally, when 50mA are achieved,

the battery charging process is considered done.

The charger is fed by a 100VA, 50Hz, 230V to

24V single phase toroidal transformer. This

transformer was chosen because the maximum

output power of the charger is 80W and, in the

majority of European countries, the characteristics of

single-phase standard electric plugs are 230V AC,

50Hz and 16A (which means 3.45kVA). The

transformer is followed by a full-bridge rectifier,

which is constituted by 4 schottky diodes MBR1660,

and a 10mF capacitive filter. This set of components

is responsible for converting the alternating voltage

ICINCO2015-12thInternationalConferenceonInformaticsinControl,AutomationandRobotics

156

(230V AC) in approximately 34V DC unregulated.

Based on these characteristics, by demanding the

maximum power, the expected output voltage ripple

is around 15V

. Due to the high transformer

impedance (when compared to the electrical grid

impedance), it was not necessary to develop a circuit

to avoid the high values of the start-up current.

A step-down DC-DC converter, working as a

pulse width modulation current-controlled voltage

source, is the power electronics converter used to

convert the input unregulated voltage (34V DC) to

an adjustable output current. The step-down DC-DC

converter switches at 1kHz. The control system is a

regular PI (proportional + integral) controller with a

sampling frequency of 4kHz, a proportional gain

(Kp) of 40 and an integral gain (Ki) of 100. Figure 2

shows the step-down model developed with PSIM

simulation software.

Figure 1: Block diagram of the charger.

In the converter shown in Figure 2, the PI controller

manages the duty-cycle applied to the MOSFET

MOS1. When the value of the output current is

lower than the desirable, the duty-cycle of the PWM

signal is increased. When the value of the output

current is higher than the desirable, the duty-cycle of

the PWM signal is decreased. The PI controller also

manages the state of the switches S1 and S2. When

it is necessary to charge the ultracapacitors module,

S1 is opened and S2 is closed. On the other hand,

when it is necessary to charge the lead-acid battery,

S1 is closed and S2 is opened. Figure 2 also shows

that the two switches (S1 and S2) are in series with

the coil L1. Because of that, the PI controller can

only change the state of the switches when the

current in the coil L1 is null. If this procedure was

not followed, it would appear an overvoltage spike

in the switch that opened. The diode D1 is the

schottky MBR1660, which is used to ensure the path

to the current when the MOSFET MOS1 (P80PF55)

is turned off. The current rating of the switches S1

and S2 is 16A. The coil L1 is used to smooth the

output current and it has 23.12mH, 428m at 1kHz.

Figure 2: PSIM model of the step-down of the charger.

2.2 The Energetically Optimized

Voltage Regulator

Figure 3 shows the block diagram of the subsystem

that uses the energetically optimized voltage

regulator. Its main goal is to manage the available

energy on the energy-storing devices, ensuring

12.5V on its output. It is considered energetically

optimized because, firstly, it uses the energy

available on the ultracapacitors module to ensure

12.5V on its output and, secondly, when there is not

enough energy on the ultracapacitors module, it uses

the energy available on both energy-storing devices

to ensure 12.5V. This architecture has the following

advantages:

 Increasing the Battery Lifespan. The fact of

using, in first place, the energy stored on the

ultracapacitors module leads to a lower battery

usage;

 Reducing the Frequency of the Battery

Replacement. The increased battery lifespan

leads to a reduction of the number of times that

the battery needs to be replaced in a specific time

period. In a long term, this fact may mean the

reduce of the cost of the system in which the

battery is installed;

 Due to the Use of Two Energy-storing

Devices, the Ultracapacitors Module can be

Fully Charged and Fully Discharged. So, there

is a better use of the ultracapacitors module

storage capacity.

To do the development, it was defined that the

mobile robot platform could demand up to 6A at

12V. Because of that, the energetically optimized

voltage regulator was designed to ensure a

maximum current value of 6A on its output. The

ANewEnergeticallyOptimizedPowerSupplySystemforaMobileRobotPlatform,usingUltracapacitorsandBatteriesto

EnsureBothUltra-fastChargingandAutonomy

157

reason why it was chosen 12.5V on its output

instead of 12V is related to the voltage regulator

architecture. The step-up converter does not have the

ability to block the energy flow from its input to its

output when the output voltage is lower than the

input voltage. When the battery is fully charged, its

voltage may go up to 12.5V. Therefore, using this

architecture, the way to block the energy flow (when

it is desirable) is ensuring that the output voltage is

always equal or greater than the input voltage.

Figure 3: Block diagram of the subsystem that uses the

energetically optimized voltage regulator.

The energetically optimized voltage regulator

model, developed with PSIM simulation software, is

shown in Figure 4. It is based on the SEPIC (Single-

Ended Primary-Inductor Converter) and step-up

converters. The outputs of the two converters are in

parallel and, because of that, the ground is the same

for both. The coils L1 and L2 have 42,7H and

95m at 1kHz. The coil L3 has 244H and 102m

at 1kHz. The MOSFETs MOS1 and MOS2 are the

P60NF06. The diodes D1 and D2 are the schottky

MBR1660. The capacitor C1 has 10F and the

capacitor C2 has 1mF.

When there is available energy on the

ultracapacitors module, the MOSFET MOS1 is

switched by a PI controller in order to keep the

output voltage regulated on 12.5V. In this case, the

energy only comes from the ultracapacitors module

and the MOSFET MOS2 is turned off. When the

energy on the ultracapacitors module is not enough

to keep the output voltage regulated at 12.5V, the

MOSFET MOS1 switches with 85% of duty-cycle

and the MOSFET MOS2 switches with an adjustable

duty-cycle. In this case, the energy comes from both

energy-storing devices to the output.

The computation of the PWM signal applied to

the MOSFETs MOS1 and MOS2 is based on a PI

controller with an extended output. The MOSFET

MOS1 handles a maximum duty-cycle of 85% and

the MOSFET MOS2 handles a duty-cycle value that

is given by the difference between the computed

value and 85%. For example, if the computed output

of the PI controller is 135, the duty-cycle of

MOSFET MOS1 will be 85% and the duty-cycle of

MOS2 will be 50%. The maximum duty-cycle on

each MOSFET is 85%. Therefore, the saturation

value of the PI controller is 170. The input of the PI

controller is the voltage on the output of the voltage

regulator. When its value is lower than the desirable,

the duty-cycle is increased. When its value is higher

than the desirable, the duty-cycle is decreased.

The PI controller has a proportional gain of 15

and an integral gain of 250. The sampling frequency

is about 4kHz and the switching frequency (applied

to the MOSFETs) is about 9.8kHz.

Figure 4: PSIM model of the energetically optimized

voltage regulator.

2.3 The Output Voltage Regulator

Figure 5 shows the block diagram of the subsystem

that uses the output voltage regulator. Its main

function is to supply a regulated voltage to the four

brushed DC motors of the mobile robot platform

from the 12.5V DC regulated (imposed by the

energetically optimized voltage regulator).

The output voltage regulator is based on four

independent full H-bridges. The input of each full H-

bridge is the 12.5V DC regulated and each one is

used to supply each brushed DC Motor of the mobile

robot platform. It is used a PWM signal to control

the state of each MOSFET of each bridge. The

rotation direction, the angular velocity and the

torque of each brushed DC motor are managed by a

ICINCO2015-12thInternationalConferenceonInformaticsinControl,AutomationandRobotics

158

Figure 5: Block diagram of the output voltage regulator.

MRP (Mobile Robot Platform) controller. This

controller computes, for each instant, the average

voltage that should be supplied to each motor. Based

on this calculation, it is done a conversion from the

average voltage to the duty-cycle of the PWM

signal. It is used bipolar modulation, which means

that MOSFETs work in pairs – when the duty-cycle

is 100%, the average output voltage is 12.5V and

when the duty-cycle is 0%, the average output

voltage is -12.5V.

Figure 6 presents the topology used to supply

each motor. Each motor has a rated voltage of 12V

and a maximum current of 1.5A. Because of that, it

was used two L298 integrated circuits to implement

the four full H-bridges.

Figure 6: PSIM model of the output voltage regulator.

3 SIMULATION RESULTS

It was used the PSIM simulation software to validate

all the developed blocks of the power supply system.

Using this software it was possible to analyse the

expected system performance and to prevent design

mistakes before any circuit implementation, as well

as making adjustments to the gains of the PI

controllers to achieve a better overall system

performance. The most important simulations results

are presented in this section.

3.1 The Charger

The charger was simulated when it is feeding the

ultracapacitors module. A maximum output power

of 80W, a maximum output current of 14A and a

final voltage of 15.5V were considered. Figure 7

shows that the ultracapacitors module charging

process is performed successfully. It is also shown

that, when the ultracapacitors module is charged, the

system begins to charge the lead-acid battery.

Figure 7: PSIM simulation results of the charger when it is

charging the ultracapacitors module up to 15.5V.

3.2 The Energetically Optimized

Voltage Regulator

The energetically optimized voltage regulator was

simulated supplying a 12.5V, 1A load. It was

considered that the ultracapacitors module and the

battery were fully charged. In Figure 8 it is noticed

that the circuit draws more current from the

ultracapacitors module as its voltage drops, in order

to keep constant the output power. Furthermore, it is

shown that the energetically optimized voltage

regulator is able to take energy from the battery

when the voltage on the ultracapacitors module

drops too much and it is able to take almost all the

energy available on the ultracapacitors module.

3.3 The Output Voltage Regulator

The output voltage regulator was simulated

considering an input voltage of 12.5V (it is the

output of the energetically optimized voltage

regulator), a duty-cycle of 75% and an inductive

load of 6 and 500H. This simulation is important

to check the voltage and current behaviour when the

output voltage regulator is supplying a DC brushed

motor (it is an inductive load with electromotive

force). In Figure 9 it is visible that the output voltage

is a square ware, with a maximum of 12.5V and a

minimum of -12.5V, where the average output

voltage depends of the duty-cycle. In spite of the

usage of an output square wave (it is constantly

ANewEnergeticallyOptimizedPowerSupplySystemforaMobileRobotPlatform,usingUltracapacitorsandBatteriesto

EnsureBothUltra-fastChargingandAutonomy

159

Figure 8: PSIM simulation results of the energetically

optimized voltage regulator when it is providing 12.5V,

1A on its output.

changing), there are no problems to DC brushed

motors because current is filtered by its inductance

and switching intervals are much shorter than the

mechanical time constants involved.

Figure 9: PSIM simulation results of load voltage and

current when the output voltage regulator has a duty-cycle

of 75% and an input voltage of 12.5V.

4 SYSTEM IMPLEMENTATION

Section 4 presents a detailed description of the real

system implementation. It was done after having

simulated and validated all the system architecture

by using the PSIM simulation software.

4.1 Hardware

The charger is able to charge both energy storing

devices. It charges the ultracapacitors module in first

place and then it charges the lead-acid battery. This

strategy was followed so as to ensure ultra-fast

charging (reducing at minimum the charging time).

The charger hardware was fully mounted onboard

the mobile robot platform to ensure that whenever it

is necessary to charge, no external additional

hardware is needed – Unlike a previous work by

Arantes et al., (2014), the charger does not need of a

dedicated charging station. The hardware

implementation was made by developing a printed

circuit board (PCB) using the CadSoft Eagle 6.4.

The energetically optimized voltage regulator is

capable of providing 12.5V on its output from the

energy available on both energy storing devices. Its

PCB implementation can be seen in Figure 10. That

Figure also shows the final implementation of the

whole power supply system: the charger, the

energetically optimized voltage regulator and the

output voltage regulator. The output voltage

regulator is able to provide the suitable average

voltage for each brushed DC motor ensuring that the

mobile robot platform moves to where is required.

That voltage is computed by the MRP controller,

which is based on the Atmel ATmega 1284P

microcontroller.

Figure 10: Final version of the mobile robot platform. It is

visible the whole power supply system: the charger PCB

(1), the energetically optimized voltage regulator PCB (2)

and the output voltage regulator PCB (3).

4.2 Software

The charger is capable of charging both energy

storing devices. For each time instant, the operation

mode of the charger is dependent of the energy

storing device that is being charged. The output

current and voltage of the charger are managed by

the algorithm that was described in Section 2.

The energetically optimized voltage regulator

uses a non-conventional digital PI controller, which

was implemented on another Atmel ATMega 328P

microcontroller. Its main particularity is the fact that

it has an extend output to deal with both SEPIC and

step-up converts. According to the algorithm, for a

computed duty-cycle equal or lower than 85%, it is

switched only the SEPIC converter. For higher duty-

ICINCO2015-12thInternationalConferenceonInformaticsinControl,AutomationandRobotics

160

cycle values, both converters are switched having in

account that, for each converter, the duty-cycle value

can never be higher than 85%. If that precaution was

not taken, due to the nature of the circuit, the power

converters could be destroyed by a short circuit.

The output voltage regulator takes the 12.5V

from the energetically optimized voltage regulator

and provides the suitable average voltage for each

brushed DC motor of the mobile robot platform. It

was implemented a digital PID controller to ensure

that each wheel of the mobile robot platform has a

measured velocity that closely follows a reference

velocity specified by the MRP controller, whose

main function is to define the trajectories of the

mobile robot platform. The algorithm of the output

voltage regulator was implemented on an Atmel

ATmega 1284P microcontroller.

All the Atmel ATmega microcontrollers used on

the power supply system were programmed using

the Atmel Studio 6.1 software.

5 EXPERIMENTAL RESULTS

Some experimental tests were made to ensure the

correct operation of each component of the system

and to further validate the proposed topology.

5.1 The Charger

The charger module was tested by charging the two

energy-storing devices. By considering only

theoretical calculations, the expected charging time

for the ultracapacitors module was, approximately, 3

minutes and 20 seconds. Because the charging

current is limited to a maximum of 14A, the

maximum charging power (80W) was only achieved

when the voltage in the ultracapacitors module

reached 5.7V. The practical results showed that the

real charging time is 3 minutes and 42 seconds. The

cause of the time gap between theoretical and

practical results is the fact that theoretical

calculations did not consider all the circuit losses

and parasitic effects. Some of them are very hard

and require very expensive equipment to be

assessed. The ultracapacitors module charging was

considered done when its voltage reached 15.5V.

After that, the lead-acid battery charging proceeded.

The measured charging time for the battery was

approximately 8 hours and 30 minutes.

5.2 The Energetically Optimized

Voltage Regulator

The main function of this voltage regulator is to

ensure 12.5V at its output using the energy available

on both the ultracapacitors module and the battery

by extracting energy from both of them. For an

ultracapacitors module voltage equal or higher than

2.48V, the voltage regulator only uses the energy

available on the ultracapacitors module. For an

ultracapacitors module voltage lower than 2.48V,

the system uses the energy available on both energy

storing devices.

It was also found that the energetically optimized

voltage regulator is able to extract almost all the

energy available on the ultracapacitors module,

taking its voltage almost to 0V.

5.3 The Output Voltage Regulator

The output voltage regulator is the last stage of the

power supply system of mobile robot platform.

Considering both energy storing devices fully

charged and running some experimental tests, it was

found that, by placing the mobile platform on a flat

surface at maximum speed (approximately 60 cm/s),

the average autonomy was 23 minutes and 13

seconds using only the available energy on the

ultracapacitors module. Using all the available

energy on both sources it was achieved an autonomy

of approximately 9 hours and 40 minutes.

6 CONCLUSIONS AND FUTURE

DEVELOPMENTS

An electrical power supply system for a mobile

robot platform has been developed. The system uses

ultracapacitors and batteries as energy-storing

devices. This paper described a major improvement

of a previous work done by Arantes et al., (2014), as

the new system is much lighter and more compact,

which allowed all of its components to be mounted

onboard the mobile robot platform. The system is

composed of three modules: 1) a charger and the

energy-storing devices; 2) an energetically

optimized voltage regulator; 3) an output voltage

regulator.

With regard to the charger, the charging time

measured for the ultracapacitors module was,

approximately, 3 minutes and 42 seconds (starting

with a fully-discharged module). Because the

charging current is limited to a maximum of 14A,

ANewEnergeticallyOptimizedPowerSupplySystemforaMobileRobotPlatform,usingUltracapacitorsandBatteriesto

EnsureBothUltra-fastChargingandAutonomy

161

the maximum charging power (80W) was only

achieved when the ultracapacitors module voltage

reached 5.7V. The ultracapacitors module charging

was completed when its voltage reached 15.5V.

After that, the lead-acid battery charging proceeded.

A small disturbance is visible in the voltage and

current waveforms at the output of the energetically

optimized voltage regulator, considering a power of

12.5W. This is due to the starting of the step-up

converter. Until then, only the ultracapacitors

module was providing power to the load. Observing

the voltage and current waveforms in the

ultracapacitors module, it is possible to conclude

that the average current value tends to increase due

to the decreasing in the average voltage value, as

expected, because only then is it possible to maintain

a constant output power.

Regarding the practical implementation of the

two previous systems, printed circuit boards were

implemented using the Eagle Cadsoft 6.4 and digital

PI controllers were developed with Atmel Studio 6.1

and microcontrollers Atmel ATmega 328P.

The output voltage regulator is the last stage of

the power supply system. It is responsible for

supplying the suitable average voltage to the motors

of the mobile robot platform. Autonomy tests were

made and it was found that, if the mobile robot

platform is placed on a flat and horizontal surface at

maximum speed (about 60 cm/s), the average

autonomy is about 23 minutes and 13 seconds using

only the available energy in ultracapacitors module.

From that moment, which corresponded to an

ultracapacitors module voltage of 2.48V, the system

started to use the energy available in both sources

and achieved an autonomy of, approximately, 9

hours and 40 minutes. It was verified that the system

was able to extract almost all of the energy stored in

the ultracapacitors module, taking its voltage almost

to 0V. Regarding the practical implementation of

this last system, a printed circuit board was also

developed through the Eagle Cadsoft 6.4, and a

digital PID controller was implemented using the

Atmel software Studio 6.1 and the microcontroller

Atmel ATmega 1284P.

As a future development, it is suggested to

modify the circuits and the control systems in order

to allow energy interchange between the

ultracapacitors module and the batteries. This will

improve energy efficiency in the event that only a

fraction of the energy stored in the ultracapacitors

module is used.

ACKNOWLEDGEMENTS

This work has been supported by FCT – Fundação

para a Ciência e Tecnologia in the scope of the

project: PEst-UID/CEC/00319/2013.

REFERENCES

Arantes, C., Sepúlveda, J., Esteves, J., Costa, H., Soares,

F., Using Ultracapacitors as Energy-storing Devices

on a Mobile Robot Platform Power System for Ultra-

fast Charging; ICINCO 2014, 11th International

Conference on Informatics in Control, Automation and

Robotics, 2014, Page(s): 156 – 164, ISBN 978-989-

758-040-6. DOI: 10.5220/0005061801560164.

Awerbuch, J. J., Sullivan, C. R., Filter-based Power

Splitting in Ultracapacitor-Battery Hybrids for

Vehicular Applications; COMPEL 2010, Control and

Modeling for Power Electronics, 2010, Pages(s): 1 –

8. DOI: 10.1109/COMPEL.2010.5562429.

Bernholc, J., Ranjan, V., Zheng, X. H., Jiang, J., Lu, W.,

Abtew, T. A., Boguslawski, P., Nardelli, M. B.,

Meunier, V., Properties of High-Performance

Capacitor Materials and Nanoscale Electronic

Devices; HPCMP-UGC 2010, High Performance

Computing Modernization Program Users Group

Conference, 2010, Page(s): 195 – 200. DOI:

10.1109/HPCMP-UGC.2010.76.

Haifeng, Dai and Xueyu, Chang; A Study on Lead Acid

Battery and Ultra-capacitor Hybrid Energy Storage

System for Hybrid City Bus; ICOIP 2010,

International Conference on Optoelectronics and

Image Processing, 2010, Page(s): 154 – 159. DOI:

10.1109/ICOIP.2010.321.

Kularatna, N., Patel, A., Supercapacitors for Energy

Mangement in Autonomous Sensor Nodes; WAC,

World Automation Congress, 2014, Page(s): 604 –

609. DOI: 10.1109/WAC.2014.6936064.

Monteiro, J., Garrido, N., Fonseca, R., Efficient

Supercapacitor Energy Usage in Mobile Phones; ICCE

2011, 2011 International Conference on Consumer

Electronics – Berlin, 2011, Pages(s): 318 – 321. DOI:

10.1109/ICCE-Berlin.2011.6031796.

Muffoletto, D., Mandris, C., Olabisi, S., Burke, K.,

Zirnheld, J., Moore, H., Singh, H., Design and

Analysis of a Smart Power Management System for

Ultracapacitor-Powered Robotic Platform; IPMHVC,

Power Modulator and High Voltage Conference,

2010, Page(s): 643 – 646. DOI:

10.1109/IPMHVC.2010.5958441.

Musat, A. M., Carp, M., Borza, P., Musat, R., Sojref, D.,

Hybrid Storage Systems and Dynamic Adapting

Topologies for Vehiche Applications; OPTIM 2012,

2012 13th International Conference on Optimization

of Electrical and Electronic Equipment, 2012, Page(s):

1842 – 1566. DOI: 10.1109/OPTIM.2012.6231910.

Niemoeller, B. A. and Krein, P. T., Battery-Ultracapacitor

ICINCO2015-12thInternationalConferenceonInformaticsinControl,AutomationandRobotics

162

Active Parallel Interface with Indirect Control of

Battery Current; PECI 2010, Power and Energy

Conference at Illinois, 2010, Pages(s): 12 – 19. DOI:

10.1109/PECI.2010.5437163.

Petchjatuporn, P., Wicheanchote, P., Khaehintung, N.,

Kiranon, W., Sunat, K., Chiewchanwattana, S.,

Intelligent ultra-fast charger for Ni-Cd batteries;

ISCAS 2005, IEEE International Symposium on

Circuits and Systems, 2005, Page(s): 5162 – 5165 Vol.

5. DOI: 10.1109/ISCAS.2005.1465797.

Qin, G., Zhu, H., Hybrid Power Supply of Rescue Robot

Based on Supercapacitors; ICMACE 2011, 2th

International Conference on Mechanic Automation

and Control Engineering, 2011, Page(s): 1350 – 1353.

DOI: 10.1109/MACE.2011.5987194.

Schneuwly, A., Gallay, R., Properties and Applications of

Supercapacitors from the State-of-Art to Future

Trends; Proceeding PCIM 2000, 2000.

Siguang, Li, Chengning, Zhang, Shaobo, Xie, Research

on Fast Charge Method for Lead-Acid Electric

Vehicle Batteries; ISA 2009, International Workshop

on Intelligent Systems and Applications, 2009,

Page(s): 1 – 5. DOI: 10.1109/IWISA.2009.5073068.

ANewEnergeticallyOptimizedPowerSupplySystemforaMobileRobotPlatform,usingUltracapacitorsandBatteriesto

EnsureBothUltra-fastChargingandAutonomy

163