Motorcycle Riding Simulator Controllability and Simulator Sickness:

A Proof-of-Concept System

Pauline Michel

, Stéphane Espié

and Samir Bouaziz

Université Paris-Saclay, ENS Paris-Saclay, CNRS, SATIE, 91190, Gif-sur-Yvette, France

TS2-SATIE-MOSS, Univ. Gustave Eiffel, IFSTTAR, F-77454 Marne-la-Vallée, France

Keywords: Driving Simulator, Simulator Sickness, Sensory Fidelity, Hardware/Software Control, Human-simulator

Interaction, Sensor-actuator Synchronization.

Abstract: Driving a motorcycle relies on the feedback provided by several human sensory systems, on the one hand,

and anticipation of the consequences of control actions, on the other hand. Driving simulators aim to create

the illusion of driving by stimulating the driver’s sensory systems. However, a significant number of drivers

experience simulator sickness, which hinders the usefulness of driving simulators in their applications, such

as driving behavior research or training / re-training. Simulator sickness occurrence is often attributed to

sensory conflict. In this work, we propose an approach to understanding simulator sickness by considering

the need for coherence between the complexity of the vehicle model and the complexity of the simulator from

a hardware point-of-view, which constrains the fidelity of the reproduced sensory stimuli. We then describe

the design of a proof-of-concept system that considers the particular issue of haptic feedback for the

handlebars of a motorcycle-riding simulator. We will use this system in further experiments to demonstrate

the impact of the coherence or mismatch of those two aspects on controllability and simulator sickness

occurrence.

1 INTRODUCTION

Driving a vehicle requires the use of several human

sensory systems: the visual, vestibular, haptic, and

auditory are the main ones. Each of them plays a

different role in the accomplishment of the driving

task. In combination, they make it possible to

estimate distance and speed, crucial parameters for

driving, particularly for vehicle trajectory control,

e.g. braking or collision avoidance. The coherence of

the various sensory feedback, and the removal of any

ambiguity between them, are ensured thanks to

multisensory integration, i.e. fusion of this

information carried out by the brain. The resulting

information is the movement of the body in relation

to the vehicle and its environment. In a driving task,

this allows the driver to decide on a single

interpretation of the current state of the vehicle being

driven (position, speed, acceleration) (Kemeny et al.,

2020), as well as the current state of other objects in

https://orcid.org/0000-0002-9743-2402

https://orcid.org/0000-0002-3449-8279

https://orcid.org/0000-0002-6768-1723

the environment (e.g., other vehicles, pedestrians,

road markings and signs, etc.).

The driver's sensory feedback is not sufficient to

accomplish a driving task. The vehicle must also be

guided to the desired destination (short-term and

medium-term), which is an active closed-loop

control-command and guidance task. The driving

activity is traditionally divided into three task levels:

strategical, tactical, and operational (Michon, 1985;

see also Motte et al., 2019). To control the vehicle

efficiently, i.e. to converge towards precise control,

with minimal oscillations, a model of the controlled

vehicle is required. In the case of human motor

control, it is called the internal model and is learned

and reinforced by experience (Wolpert et al., 2011;

McNamee & Wolpert, 2019; Pierella et al., 2019).

The goal of a driving simulator is to create the

illusion of driving by stimulating the driver's sensory

systems (Siegler et al., 2001; Fischer et al., 2016;

Salisbury & Limebeer, 2017). Inevitably, driving

406

Michel, P., Espié, S. and Bouaziz, S.

Motorcycle Riding Simulator Controllability and Simulator Sickness: A Proof-of-Concept System.

DOI: 10.5220/0010576704060413

In Proceedings of the 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2021), pages 406-413

ISBN: 978-989-758-528-9

simulators provide only a subset of the sensory

stimuli available in a real driving situation. The

implementation of a high-fidelity dynamic vehicle

model is a design constraint considered necessary to

best reproduce real-world driving situations. This

constraint, called physical validity (Malaterre &

Fréchaux, 2001; see also Faure, 2017, and Lobjois et

al., 2021), is often taken into account without concern

for the hardware architecture of the simulator and the

sensory cues' fidelity. However, the poor or delayed

restitution of a stimulus, similarly to the absence of

its restitution, can lead to multisensory integration

failing. Furthermore, all drivers do not use each

sensory feedback in the same way. This makes it

difficult to build a simulator suitable for all drivers

and all purposes. Multisensory integration relies on

the redundancy of sensory signals for disambiguation,

but this is not always possible in driving simulators.

This causes discomfort for drivers because it results

in sensory conflict, which is identified and widely

accepted as a cause of Simulator Sickness (SS) as

well as Motion Sickness (MS) (Reason & Brand,

1975).

In this work, we propose an approach to

understanding SS, an adverse physiological reaction

to a simulated driving situation. While MS primarily

affects passengers in vehicles, all users of driving

simulators can suffer from SS (Diels, 2016; Iskander

et al., 2019). This is an essential concern for all

driving simulator applications. We argue that SS

comes from inadequacy between the complexity of

the vehicle model and qualities of the cues provided

to the driver (*). Driving simulators often implement

a high complexity vehicle model, e.g. with a large

number of degrees of freedom and/or non-linearities,

etc. Using a high complexity vehicle model, a driving

simulator can simulate the dynamics of the real

vehicle with high fidelity. However, if the

architecture (HW/SW) of the simulator is not adapted

accordingly, all of the sensory cues corresponding to

the modeled physics cannot be provided to the driver.

For example, in the case of motorcycle riding, the

gyroscopic effect can be modeled but cannot be

rendered using current technology. This inadequacy

may lead to uncontrollability or poor controllability

of the simulated vehicle that induces SS.

We focus on the complex case of reproducing

haptic cues on the handlebars of a motorcycle driving

simulator. We then describe the design of our Proof-

Of-Concept (POC) system, which we plan to use to

test our hypothesis (*) in various experiments. Our

POC system is a motorcycle driving simulator with

haptic feedback on the handlebars. Its design takes

into account the specific constraints of human sensory

systems.

2 ARCHITECTURE / MODEL

MISMATCH AND SIMULATOR

SICKNESS: OUR HYPOTHESIS

Symptoms of SS may vary in type and intensity

depending on the individual (Schweig et al., 2018).

They can be separated into three clusters of

symptoms: (1) oculomotor symptoms, (2)

disorientation, and (3) nausea (Kennedy et al., 1993).

Symptoms and intensity may also vary across

situations for the same individual. In particular, they

depend on the simulated task to accomplish

(Kolasinski, 1995): for example, cornering is one of

the most SS-inducing tasks, especially for the

smallest curvature radii. Rich environments, such as

urban junctions, are also particularly problematic.

However, these situations cannot be excluded from

driving simulators experiments. They are of great

interest both in road safety research and public

education initiatives and in training/retraining

applications.

Experiencing symptoms of SS can affect the

driver’s task performance and/or reduce experiment

duration (Money, 1970; Stoner et al., 2011; Liebherr

et al., 2020). Besides, when a susceptible driver does

not quit the experiment of their own accord,

experiments are often only stopped when the

experimenter is forced to (e.g., after the driver

vomits). This constitutes a bad experience for the

driver and can lead to them having a negative

preconception of driving simulators in general. If the

person agrees to participate in further driving

simulator experiments, this tainted image can produce

anxiety and pre-discomfort (Liebherr et al., 2021),

which have been demonstrated to be negatively

linked to SS (Bertin et al., 2004; Stelling et al., 2021).

This negative preconception can also lead to

definitive refusal of further participation in such

experiments. Furthermore, the elderly population is

particularly susceptible to SS. All of this induces

involuntary “at the door” filtering of the population

studied on driving simulators, i.e. the selection of the

population that is not susceptible to SS. Recent

studies report drop-out rates due to simulator sickness

ranging from 5% to 30% when the participants are

part of the general population (Balk et al., 2013;

Liebherr et al., 2020; Saredakis et al., 2020).

Additionnally, Matas et al. (2015) reported a drop-out

rate of 59% for an experiment focusing on older

Motorcycle Riding Simulator Controllability and Simulator Sickness: A Proof-of-Concept System

407

adults. The results acquired thanks to driving

simulators are hence often biased.

In a simulated driving simulation, as opposed to a

real-world driving situation, the driver teleoperates a

vehicle model, as represented in Figure 1. This is

fundamentally different from driving a vehicle.

Teleoperation control-command rules could be used.

Figure 1: Illustration of the interactions between driver and

driving simulator.

Motorcycle riding is more complex than driving a

car. The rider controls the trajectory of their

motorcycle through two torques: the roll torque and

the steering torque, i.e. the torque applied by the rider

on the motorcycle handlebars. The handlebars of a

motorcycle serve a double action-perception purpose:

the rider controls the system by interacting with them

and they provide sensory feedback. Haptic cues on

the handlebars are essential to the rider and

significantly affect their riding behavior. For

example, they provide feedback on the interaction

between the motorcycle’s tires and the road.

However, measuring the rider’s steer torque is a

complex issue: when the rider exerts a torque on the

motorcycle’s handlebars, the steering column rotates.

Moreover, at high speed, the variations of the angular

position are of small amplitude. This means that any

torque measure will not only reflect the torque

applied by the rider but also the motorcycle’s inherent

dynamics. In motorcycle riding simulators, restitution

of the sensory cues corresponding to haptic

perception on the handlebars is therefore particularly

complex and crucial. Poor or delayed restitution of

haptic cues hinders the controllability of the virtual

vehicle.

Motorcycles are inherently dynamically unstable:

a rider needs to stabilize their motorcycle to ride it.

That is why controllability is a crucial concern for

motorcycle riding simulators, and similarly for car

driving simulators. However, the research on the link

between the controllability of a simulator and SS is

still limited. Car-driving and motorcycle-riding

simulators are currently used only in situations where

they are fully controllable. However, as we

previously mentioned, this means that driving

simulation usage is deprived of situations that are of

great interest such as driving at urban junctions.

A simulator being non-controllable may result in

erratic, oscillating movements that produce

uncontrolled image rotations, which have been shown

to cause SS occurrence (Golding, 2006; Cohen et al.,

2019). Moreover, experiencing control difficulties

may prompt the driver to feel anxious and

uncomfortable, feelings which, as mentioned above,

are also negatively linked to SS.

Our opinion is that a mismatch between the

complexity of the vehicle model and the fidelity of

the sensory stimuli that correspond to it prevents the

driver from being able to adequately control the

virtual vehicle, which then induces SS. As discussed,

SS occurrence and SS symptoms severity are

intrinsically linked to psychological validity, which

we believe should be the goal in designing driving

simulators for road safety research or training

applications. In the following, we focus on the design

of a POC system for this hypothesis, with a special

interest in providing haptic feedback.

3 OUR PROOF-OF-CONCEPT

SYSTEM

Because driving a motorcycle involves several human

sensory systems, each of which the precise role in the

driving task depends on the rider, deciding what

sensory stimuli is reproduced and how is a complex

issue. However, human sensory systems have

inherent time and frequency sensitivities, as well as

physical and biochemical limitations that need to be

taken into account. For example, sensory receptor and

neuromuscular dynamics, nerve conduction, and

neural processing altogether are responsible for a time

delay between the instant when a sensory stimulus (or

stimuli) is applied and the instant when the control

response begins. Time delays respectively introduced

by the visual and haptic system are presented in

Table 1. In this section, we describe the design of our

POC system under these constraints.

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

408

Table 1: Sensory delays characteristic of the visual and

haptic systems (Nash et al., 2016).

Sensory system Sensory dela

Visual 100ms

–

560ms

Haptic

> 34ms

or > 48ms

(depending on

the rece

tors

)

3.1 System Architecture

Our POC system, represented in Figure 2 provides the

driver with haptic cues using motorcycle handlebars

mechanically coupled to a CanisDrive-20A-160-AM-

H-SIE servo actuator, pictured in Figure 3. Our

system also provides visual cues using a Virtual

Reality (VR) headset, the HTC Vive Pro system. The

goal for the driver is to control the trajectory of a

virtual motorcycle through a virtual scene using this

bimodal feedback.

Figure 2: Schematic representation of the interactions

between the driver and our POC system.

Figure 3: Haptic feedback motorcycle handlebars used in

our POC system.

We use a distributed architecture, as represented

by its deployment diagram in Figure 4. It is

implemented across:

 a PC embedding an Intel® Core™ i7-8700

CPU @ 3.20GHz and a Nvidia GeForce GTX

1060 responsible for the computing of the

dynamic model of the simulated system (e.g.,

bicycle or motorcycle). The dynamic model is

computed at a frequency of 1 kHz. This PC is

also responsible for generating the images for

visual rendering;

 a lab-made board embedding a mbed

(LPC1768) microcontroller and a Field-

Programmable Gate Array (FPGA), a DE0-

Nano board, responsible for the data

acquisition of the torque applied to the

handlebars by the driver;

 a lab-made board embedding a STM32F446

microcontroller and a DE0-Nano board,

responsible for generating haptic cues in

conjunction with a YukonDrive-1021-ADO

servo controller.

This distributed architecture guarantees the stability

of the calculation, sampling, and transmission

frequencies imposed by the constraints we will

discuss in this part. Frequency jitters would

negatively impact the controllability of the riding

simulator, which we want to avoid per our hypothesis

(*).

Figure 4: Deployment diagram of our POC system.

The servo actuator and servo controller are

isolated with special care to avoid electromagnetic

interference between them and the VR headset’s

display.

3.2 Human Control Input

In motorcycle riding simulation, as previously stated,

the rider does not drive a real motorcycle, but

remotely operates a motorcycle model by interacting

with a physical system. In the case of our POC

system, the driver controls the trajectory of the

motorcycle model by acting on the handlebars. The

resulting torque on the steering column is an input of

the motorcycle model and needs to be measured. The

Motorcycle Riding Simulator Controllability and Simulator Sickness: A Proof-of-Concept System

409

servo actuator we use has been specifically picked

because of its high gear ratio (R=160), which makes

it non-manually reversible, ensuring the separation of

the system’s inherent dynamics and the human action

on the system. The torque exerted by the driver on the

handlebars can thus be measured directly using a

strain gauge.

The torque data acquisition board was designed so

that the embedded FPGA can sample the measure of

up to four sensors via SPI communication. In our

application, only one channel is used. The sampled

measures are sent to the embedded microcontroller

after being requested, also via SPI communication.

Both SPI blocks implemented on the FPGA use a

10MHz SCLK frequency for data transfer. The

microcontroller filters the strain gauge data using an

Infinite Impulse Response (IIR) filter, specifically a

order low-pass Butterworth numerical filter. The

filter data is then re-transmitted via CAN bus, with a

CAN bus speed of 1Mbits/s.

3.3 Visual Cueing

Visual cues projected into the Head-Mounted Display

(HMD) are computed in real-time by a 3D graphics

generator, the Unity3D engine with the OpenXR

plugin. Relevant model outputs (e.g., position, speed,

acceleration) are sent over Ethernet using UDP at a

frequency of 90 Hz. Using an HMD will also allow

us to implement audio cueing in further work.

For visual rendering, the first time-related

constraint that was taken into account is the image

refresh rate. A constant, sufficient refresh rate is

necessary for the driver to operate under the

impression of continuous, fluid visual motion. 30

frames per second (FPS) is commonly defined as the

acceptable minimum frame rate for this purpose.

However, the images projected in the case of driving

simulators often include vehicles moving at high

speeds, which require a higher refresh rate for the

animation to appear continuous and fluid. Moreover,

a variable refresh rate induces image flickering, as

well as instability of the virtual environment when

using an HMD. These visual effects result in erratic,

oscillating movements of the simulated motorcycle

from the point-of-view of the driver. In our system,

visual cues are generated by the 3D graphics

generator at a constant refresh rate of 90FPS.

As presented in Table 1, the visual system is

characterized by a sensory delay between the

perception of a visual change in the environment and

control response ranging from 100ms to 560ms (Nash

et al., 2016). However, drivers are sensible to much

lower transport delay, i.e. time difference between the

instant of a control-command action – in the case of

our system, turning the handlebars – and system

response. For vision, system response is a change in

the visual scene. When the transport delay introduced

by a system is greater than an acceptable transport

delay, the system becomes more difficult to control or

even uncontrollable. It also causes uneasiness for the

driver. In our application, feedback is bimodal:

visual-tactile. This impacts acceptable latencies for

both the visual and haptic feedback. The maximum

acceptable transport delays are system-, task- and

person-dependent (Attig et al., 2017). For a simulated

driving task, which is a time-critical task, the

maximum acceptable visual latency reported in the

literature is 50ms (Frank et al., 1988; Padmos &

Milders, 1992). As previously stated, in our system,

the VR environment simulation runs at 90FPS.

Assuming that all computations are performed within

one frame, this frame rate alone introduces a latency

of 11ms. However, there are additional software and

hardware sources of transport delay. For the HTC

Vive Pro used in conjunction with the Unity3D game

engine, Le Chénéchal and Chatel-Goldman (2018)

found a mean transport delay of 31.33ms. This is an

acceptable visual latency for a simulated driving task.

However, transport delay does not only depend on the

visual rendering sub-system, but also on the human

haptic cueing sub-system, which will be discussed in

the next subsection.

In conclusion of this section, the design of the

visual cueing sub-system of our POC system takes

into account physiological constraints specific to

vision in the particular context of having to

accomplish a simulated driving task. Consideration of

these constraints allows us to avoid involuntary non-

controllability of the simulated motorcycle and image

flickering and/or oscillations. This will enable us to

test our hypothesis (*) by changing visual feedback

modalities without the risk of uncontrolled changes in

those modalities caused by the system.

3.4 Haptic Cueing

Haptic perception is divided into two dependent

sensory sub-modalities (Reed & Ziat, 2018):

 kinesthesia, i.e. the perception of the body’s

movement thanks to proprioceptive sensors

that provide feedback on efforts endured by

the muscles and on the angular position of the

body’s limbs;

 tactile perception, or sense of touch, i.e., the

perception of the skin’s interaction with the

environment (pressure, vibration, temperature,

texture, roughness, etc.) thanks to cutaneous

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

410

surface feedback on the material properties of

objects in contact with the body and the

angular position of its limbs.

The mechanoreceptors, specific sensory receptors

located in the different layers of the skin and the joints

and muscles, are respectively responsible for tactile

perception and kinesthesia. There are several types of

mechanoreceptors, which are each sensitive to

specific stimuli of different frequencies. Riding a

motorcycle using handlebars activates three

particular types of mechanoreceptors, the

characteristics of which are summarized in Table 2.

Their respective frequency sensitivities impose

design constraints for the frequency-related

parameters of haptic cueing in our POC system. The

torque exerted by the driver on the handlebars is

sampled at a frequency of 500Hz. This is coherent

with the frequency sensitivities of the

mechanoreceptors involved in the driving task. It also

complies with teleoperation control-command rules

that specify a haptic refresh rate in the range of 500Hz

to 1kHz to ensure the stability and transparency of the

haptic interaction. This loop frequency guarantees the

controllability of the system and thus of the simulated

motorcycle.

The haptic rendering board was specifically

designed for this application so that the embedded

FPGA can sample the encoder data directly from its

serial transmission by the servo controller. Like the

strain gauge data acquisition board, the FPGA and

microcontroller of this board communicate using SPI

with a 10MHz SCLK frequency. Encoder data, i.e.

position and speed of the handlebars, is re-transmitted

over CAN bus at a 1ms period by the microcontroller.

This information is used as inputs of the dynamic

model. This sampling and transmission frequency is

compliant with the haptic loop constraints previously

established.

Table 2: Characteristics (stimulation type sensitivity and

frequency sensitivity) of the mechanoreceptors relevant to

the driving task (Hale & Stanney, 2004).

Mechanoreceptors

Stimulation type

(relevant for the

driving task)

sensitivit

Frequency

sensitivity

Pacinian corpuscles

Vibration,

acceleration

100Hz –

1kHz

Ruffini endings

lateral force,

motion direction,

static force

0.4Hz –

100Hz

Meissner corpuscles

Velocity, grip

control

2Hz –

40Hz

We have described what feedback is transmitted

to the PC that computes the dynamic model, and how

and when it is. Our haptic feedback actuator is speed

controlled using a Proportional Integral (PI)

controller. However, in our implementation, the

speed output of the model is not applied directly as

speed reference but is first corrected using the current

position error. This position-speed dual control

avoids position derivation due to incremental speed

tracking error and numerical integration. The block

diagram representation of the position-speed control

is given in Figure 5. For clarity, model inputs are not

exhaustively represented, but they have been

described in the text.

Figure 5: Block diagram representation of the servo

actuator’s position-speed control.

Similarly to vision, haptic perception allows for a

maximum acceptable transport delay between the

driver trying to turn the handlebars and the handlebars

effectively turning. Even though nerve messages have

a longer distance to travel between the arms and

hands and the brain than between the eyes and the

brain, delays characteristic of haptic perception are

significantly shorter than delays characteristic of

vision (as evidenced in Table 1; see also Cameron et

al., 2014; Crevecoeur et al., 2016). The maximum

acceptable haptic delay varies significantly on the

task, the system, and the person (Kaber & Zhang,

2011). Our objective was the minimization of haptic

latency in our system. CAN bus communication

introduces a well-known delay that corresponds to the

duration of a CAN frame. The maximum duration of

a CAN base frame is around 134µs (for an 8 bytes

data frame). The CAN arbitration process also adds

another delay that is difficult to quantify. This is why

the speed reference value for the servo actuator is

transmitted via an analog input directly to the servo

controller. Speed is thus controlled while minimizing

additional delay in the system control that affects both

haptic and visual latency.

To summarize, we designed the haptic rendering

sub-system of our POC system by taking into account

physiological constraints specific to haptic

perception, such as haptic sensory receptors

frequency sensitivities and minimal haptic latency.

This ensures that this sub-system induces no

involuntary non-controllability of the simulated

Motorcycle Riding Simulator Controllability and Simulator Sickness: A Proof-of-Concept System

411

motorcycle. This will allow us to test our hypothesis

(*) by changing haptic feedback modalities without

risk of uncontrolled changes in those modalities

caused by reasons inherent to the system.

Furthermore, our POC system provides the rider with

good quality haptic feedback on motorcycle

handlebars, which we believe is necessary to ensure

the controllability of any motorcycle driving

simulator.

4 CONCLUSIONS

We argue that Simulator Sickness comes from

inadequacy between the complexity of the vehicle

model and the fidelity of the sensory cues to be

reproduced. We have taken a special interest in

motorcycle riding simulators and in particular in the

issue of providing good quality haptic feedback on

the motorcycle handlebars. Indeed, this feedback

significantly affects the simulator’s controllability

and is not often taken into account.

We aim to demonstrate the cruciality of the

coherence between both of those aspects. To do so,

we have designed a Proof-Of-Concept system that

takes into account the specific constraints of human

sensory systems. This design philosophy, detailed in

this work, will thus allow us to modulate visual and/or

haptic feedback. By doing so, we will be able to

compare the results in terms of (1) controllability and

task performance and (2) anxiety, discomfort, and

eventual SS symptoms severity of a motor control

task when the complexity of the vehicle model and

the fidelity of the sensory cues (a) when they are

coherent and (b) when they are mismatched. The

exploration of our hypothesis in the case of a “simple”

task using this POC system will be our next step. Our

haptic feedback subsystem will allow us to explore

the impact of the adequacy of the motorcycle

dynamic model’s complexity with the complexity of

the simulator architecture on trajectory control,

presence, and SS occurrence in a future experiment.

We plan to compare these aspects for coherent and

mismatched modalities defined by: (1) two dynamic

motorcycle models of different complexity, and (2)

disabled or enabled haptic restitution for the same

motorcycle riding simulator platform.

REFERENCES

Attig, C., Rauh, N., Franke, T., & Krems, J. F. (2017).

System latency guidelines then and now – is zero

latency really considered necessary? In D. Harris (Ed.),

Engineering Psychology and Cognitive Ergonomics

2017, Part II, LNAI 10276 (2-14).

Bertin, R. J. V., Guillot, A., Collet, C., Vienne, F., Espié,

S., & Graf, W. (2004). Objective measurement of

simulator sickness and the role of visual-vestibular

conflict situations: a study with vestibular-loss (a-

reflexive) subjects (poster). Driving Simulation

Conference.

Cameron, B. D., De la Malla, C., & López-Moliner, J.

(2014). The role of differential delays in integrating

transient visual and proprioceptive information.

Frontiers in Psychology, 5(50).

Cohen, B., Dai, M., Yakushin, S. B., & Cho, C. (2019). The

neural basis of motion sickness. Journal of

Neurophysiology, 121(3), 973-982.

Crevecoeur, F., Munoz, D. P., & Scott, S. H. (2016).

Dynamic multisensory integration: somatosensory

speed trumps visual accuracy during feedback control.

Journal of Neuroscience, 36(33), 8598-8611.

Diels, C., & Bos, J. E. (2016). Self-driving carsickness.

Applied Ergonomics, 53, 374-382.

Faure, V. (2017). Les simulateurs de conduite : évaluation

de la validité psychologique sous l’angle de la charge

mentale [Doctoral dissertation]. Université Paris-

Saclay.

Fischer, M., Seefried, A., & Seehof, C. (2016). Objective

motion cueing test for driving simulators. Proceedings

of Driving Simulation Conference, 41-50.

Frank, L. H., Casali, J. G., & Wierwille, W. W. (1988).

Effects of visual display and motion system delays on

operator performance and uneasiness in a driving

simulator. Human Factors, 30(2), 201-217.

Golding, J. F. (2006) Motion sickness susceptibility.

Autonomic Neuroscience: Basic and Clinical, 129(1-2),

67-76.

Hale, K., & Stanney, K. (2004). Deriving haptic design

guidelines from human physiological, psychophysical,

and neurological foundations. Proceedings of IEEE

Conference on Computer Graphics and Applications,

24(2), 33-39.

Iskander, J., Attia, M., Saleh, K., Nahavandi, D., AboBakr,

A., Mohamed, S. M. K., Asadi, H., Khosravi, A., Lim,

C., & Hossny, M. (2019). From car sickness to

autonomous car sickness: a review. Transportation

Research Part F – Traffic Psychology and Behavior,

62, 716-726.

Kaber, D. B., & Zhang, T. (2011). Human factors in virtual

reality system design for mobility and haptic task

performance. Reviews of Human Factors and

Ergonomics, 7(1), 323-366.

Kemeny, A., Chardonnet, J. R., & Colombet, F. (2020).

Self-motion perception and cybersickness. In Getting

Rid of Cybersickness (31-62).

Kennedy, R. S., Lane, N. E., Berbaum, K. S., & Lilienthal

M. G. (1993). Simulator sickness questionnaire: an

enhanced method for quantifying simulator sickness.

The International Journal of Aviation Psychology, 3(3),

203-220.

Kolasinski, E. (1995). Simulator sickness in virtual

environments (Technical Report n°1027). US Army

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

412

Research Institute for the Behavioral and Social

Sciences.

Le Chénéchal, M., & Chatel-Goldman, J. (2018). HTC Vive

Pro time performance benchmark for scientific

research. Proceedings of International Conference on

Artificial Reality and Telexistence and Eurographics

Symposium on Virtual Environments, 81-84.

Liebherr, M., Schweig, S., Brandtner, A., Averbeck, H.,

Maas, N., Schramm, D., & Brand, M. (2020). When

virtuality becomes real: relevance of mental abilities

and age in simulator adaptation and dropouts.

Ergonomics, 63(10), 1271-1280.

Liebherr, M., Mueller, S. M., Schweig, S., Maas, N.,

Schramm, D., & Brand, M. (2021). Stress and

simulated environments: insights from physiological

marker. Frontiers in Virtual Reality, 2.

Lobjois, R., Faure, V., Désiré, L., & Benguigui, N. (2021).

Behavioral and workload measures in real and

simulated driving: do they tell us the same thing about

the validity of driving simulation? Safety Science, 134.

Malaterre, G., & Fréchaux, J. (2001). Etude de la validité

des simulateurs de conduite par comparaison de tâches

réalisées en situation réelle et en simulation. In V.

Cavallo & C. Berthelon (Eds.), Facteurs perceptifs

dans les activités de transport (149-156).

Matas, N. A., Nettelbeck, T., & Burns, R. (2015). Dropout

during a driving simulator study: a survival analysis.

Journal of Safety Research.

McNamee, D., & Wolpert, D. M. (2019). Internal models

in biological control. Annual Review of Control,

Robotics, and Autonomous Systems, 2(1), 339-364.

Michon, J. A. (1985). A critical view of driver behavior

models: what do we know, what should we do? In L.

Evans & R. C. Schwing (Eds.), Human behavior and

traffic safety (485-520).

Money, K. E. (1970). Motion sickness. Physiological

Review, 50(1), 1-39.

Motte, C., Lappi, O., Giles, O., Markkula, G., Mars, F., &

Wilkie, R. (2019). Getting back into the loop: the

perceptual-motor determinants of successful transitions

out of automated driving. Human Factors, 61(7), 1037-

1065.

Nash, C. J., Cole, D. J., & Bigler, R. (2016). A review of

human sensory dynamics for application to models of

driver steering and speed control. Biological

Cybernetics, 110(2-3), 91-116.

Padmos, P., & Milders, M. V. (1992). Quality criteria for

simulator images: a literature review. Human Factors,

34(6), 727-748.

Pierella, C., Casadio, M., Mussa-Ivaldi, F. A., & Solla, S.

A. (2019). The dynamics of motor learning through the

formation of internal models. PLOS Computational

Biology, 15(12).

Reason, J. T., & Brand, J. J. (1975). Motion Sickness.

London: Academic Press.

Reed, C. L., & Ziat, M. (2018). Haptic perception: from the

skin to the brain. In Reference Module in Neuroscience

and Biobehavioral Psychology.

Salisbury, I., & Limebeer, D. (2017). Motion cueing in

high-performance vehicle simulators. Vehicle System

Dynamics, 55(6), 775-801.

Saredakis, D., Szpak, A, Birckhead, B., Keage, H. A. D.,

Rizzo, A., & Loetscher, T. (2020). Factors associated

with virtual reality sickness in head-mounted displays:

a systematic review and meta-analysis. Frontiers in

Human Neuroscience, 14.

Schweig, S., Liebherr, M., Schramm, D., Brand, M. &

Maas, N. (2018). The impact of psychological and

demographic parameters on simulator sickness. In

Proceedings of 8

International Conference on

Simulation and Modeling Methodologies, Technologies

and Applications – SIMULTECH, 91-97.

Siegler, I., Reymond, G., Kemeny, A., & Berthoz, A.

(2001). Sensorimotor integration in a driving simulator:

contribution of motion cueing in elementary driving

task. Proceedings of Driving Simulation Conference,

21-32.

Stelling, D., Hermes, M., Gerrit, H., Mittelstädt, J.,

Niedermeier, D., Schudlik, K., & Duda, H. (2021).

Individual differences in the temporal progression of

motion sickness and anxiety: the role of passengers’

trait anxiety and motion sickness history. Ergonomics.

Stoner, H., Fisher, D., & Mollenhauer, M. Jr. (2011).

Simulator and scenario factors influencing simulator

sickness. Handbook of Driving Simulation for

Engineering, Medicine and Psychology, 14-1 - 14-24.

Wolpert, D. M., Diedrischsen, J., & Flanagan, J. R. (2011).

Principles of sensorimotor learning. Nature Reviews

Neuroscience, 12(12), 739-751.

Motorcycle Riding Simulator Controllability and Simulator Sickness: A Proof-of-Concept System

413