Driver Attention in Automatic Transmission Cars

profileKouniB
DesignofAutomotiveDigitalInstrumentClusterAdjustabletoDriversCognitiveCharacteristics.pdf

Design of Automotive Digital Instrument Cluster Adjustable to Driver’s Cognitive Characteristics

Young-Jin Kwon, Jin-Kyu Choi, Juil Jeon, Kyongho Kim, Byungtae Jang Intelligent Robotics Research Division

Electronics and Telecommunications Research Institute (ETRI) Daejeon, Republic of Korea

{youngjin.kwon, jkchoi, seventhday07, kkh, jbt}@etri.re.kr

Abstract—In this paper, we propose an adjustable digital instrument cluster designed to provide an effective user interface and user experience (UI/UX) depending on the driver’s age, gender, personal driving experience, and driver’s emotions and conditions changes in driving. To make this possible, firstly we build a theme structure that can link with scripts to reconfigure flexibly the predetermined digital instrument cluster according to various situations that occur during driving, such as elderly drivers, female drivers, novice drivers, and general drivers. Secondly, we defined objects and elements to provide a digital instrument cluster that can change dynamically according to the script that makes up the cluster, and this library designed provides a work environment for users to personalize some themes. In addition, the proposed system can basic research to derive effective interaction methods by providing an environment suitable for evaluating the usefulness of each driver groups.

Index Terms—Automotive digital instrument cluster, instru- ment cluster, driver-vehicle interaction, digital cockpit, HMI

I. INTRODUCTION

Recently, many novel technologies have been applied to vehicles to provide a natural interface between humans and vehicles. The Mercedes-Benz User Experience (MBUX), the next-generation infotainment system of Daimler released at the consumer electronics show (CES) 2018, integrates digital instrument cluster and central information display (CID) using two wide liquid crystal displays (LCD). However, behind that simple appearance, Daimler’s applying new technologies such as multi-functional touch control button of the steering wheel and intelligent voice control to provide a new user experience between vehicle and driver. At the CES in the same year, Kia Motors focused on self-driving technology under the vision of “Beyond self-driving”, however at CES 2019, released “R.E.A.D: Real-Time Emotion Adaptive Driving”. The system recognizes the emotions and conditions of the driver and actively controls the system in the vehicle, such as audio, air conditioning, lighting, and steering.

Moreover, efforts are being made to provide a new human- machine interface (HMI) such as BMW’s Vision iNEXT, Byton’s massive cockpit module, and HARMAN’s digital cockpit. In this paper, we propose a digital instrument cluster system designed to provide effective UI/UX experimental environment according to the age, gender, personal experience of driving, emotion of drivers changing during operation, and condition of drivers through digital instrument clusters.

II. BACKGROUND AND RELATED WORKS

As technology applied to vehicles has increased from several years ago, the amount of information that drivers can access while driving is increasing [1]. This provides the driver with various information for safe driving. However, the information that occurs during driving causes the driver to reduce concentration and increase workload [2]. In addition, during driving, the driver acquires over 80% of the information necessary for driving through vision. This has increased the importance of in-vehicle displays, and vehicle manufacturers are focusing on enhancing the convenience and stability of drivers by providing such information [3]–[5].

The LCD-based digital instrument clusters introduced in Audi’s Virtual Cockpit or Daimler’s MBUX can represent information in a variety of ways compared to conventional analog instrument clusters or hybrid instrument clusters. How- ever, the digital instrument cluster based on a full-screen LCD has a high autonomy in the expression of information, so it can use in an ineffective method of information delivery [6], [7].

Therefore, we have developed a driver-adaptive vehicle in- teraction system (DAVIS) to provide proper interaction mech- anism according to driver’s condition or driver’s characteristic through the last research.

Moreover, we extracted the 27 user requirements of the elderly driver through a preliminary study and created 39 system requirements that are expected to be implemented in the system [8]. In [9], the basic layout of digital instrument

Fig. 1: Digital instrument cluster themes.

461978-1-7281-0893-3/19/$31.00 ©2019 IEEE ICTC 2019

Authorized licensed use limited to: Florida Institute of Technology. Downloaded on November 30,2021 at 20:54:02 UTC from IEEE Xplore. Restrictions apply.

clusters for each driver group with measured cognitive and be- havioral characteristics, such as elderly drivers, female drivers, and novice drivers was set up and guided and provided alert information in multi-modal to reduce driver workload and improve driving stability. Our previous study [9] introduces the concept design of the Digital Cockpit which varies the contents of a digital instrument cluster adaptive to the characteristics of the driver (age, gender, experience).

In this study, the profile-based digital instrument cluster structure was improved, as shown in Chapter 3, by referring to these prior research results and requirements [9], [10].

III. SYSTEM ARCHITECTURE A. User profile based theme-matching digital instrument clus- ter

We synchronize the basic information of the driver (gender, driving experience, age, etc.) in DAVIS when the driver boards the vehicle based on the information that the user profile is input to the driver’s smart device as shown in [10].

Fig. 2: Definition of instrument cluster components.

Subsequently, we evaluate the general cognitive, physical, and psychological characteristics of each driver group based on the basic information of the driver in DAVIS, and load the cluster theme suitable for the driver group in the digital instrument cluster as shown in Fig. 1. The previous study provides a cluster theme based on image resources. Therefore, there is the problem of securing image resources to change color or shape.

B. Objects Definition of Digital Instrument Cluster We define 10 major pieces of information as objects, and

each object is composed of elements, as shown in Fig. 2

Fig. 3: Predefined position layout.

for a general instrument cluster structure. A speedometer, tachometer, engine coolant temperature, fuel, warning, gear position, and odometer are indicated as essential information to provide the driver with the vehicle operating status. It provides a tire pressure monitoring system (TPMS) and advanced driver assistance systems (ADAS) information and multimedia status in the center information area, depending on the vehicle’s functional options as auxiliary information.

To access each element, we provide the Application Pro- gramming Interface (API). For example, the speedometer is composed of background, frame, tick, label, and needle. Each can load a picture file (PNG, JPG, etc.) which is an external resource and display it on the screen, or output it as the drawing object using a library API.

C. Re-configurable Theme Structure

The instrument cluster was structured using a java-script object notation (JSON) to reconstruct the theme. The JSON comprises a ’key-value’ data object in a text format that is easily readable by a human. The theme comprises three major sections as follows.

Fig. 4: Examples of the label element’s dynamic effects.

• ‘general’ section It is an item that provides basic infor- mation showing the identity of the theme, such as the unique ID of the theme, name, and registration location.

• ‘attribute’ section It is an item that includes the defini- tion of a unique layout of theme and location information for placing objects. The pre-defined position information is the same as Fig. 3.

• ‘property’ section It is an item that defines the element that makes up the object. For example, the shape of elements, detailed location (coordinate-based), size, and color of the object.

API can access all defined items, and an instrument cluster can re-configured with only changed theme script information. Fig. 4 shows an example of applying dynamic effects to a label. It is a function used to emphasize a particular section. It combines focus and scale effects to implement four actions. Fig. 5 shows examples changing the color of the elements (a) and the shapes of the indicator (b).

D. Re-configurable Theme-based System

The theme based on the structure defined above can re- configured by using the JSON script information. All elements can build objects using image resources, or they can realize objects using purely drawing API. This system is based on the Android platform and can be implemented in the application. To enhance portability, the instrument cluster graphics library

462

Authorized licensed use limited to: Florida Institute of Technology. Downloaded on November 30,2021 at 20:54:02 UTC from IEEE Xplore. Restrictions apply.

(a) the color of elements (b) the shapes of an indicator

Fig. 5: Examples about changing the color of elements and changing the shape of an indicator element.

is modularized. Fig. 6 shows an example of the operation of the instrument cluster based on the JSON theme structure designed above. This works represented to the image-based gauges and drawing-based gauges. The drawing objects can have more effects than image objects. It can configure as a fuel gauge (Hybrid object) shown in Fig.6, it can be configure using a mix of image resources (Background and Tick Marker) and the drawing objects (Indicator and Label).

Fig. 6: Examples of the adjustable digital instrument cluster by theme script.

IV. CONCLUSION Recently, automobile cockpit environment is changing from

the classical method using analog instrument clusters to full- screen digital instrument clusters that provide drivers with diverse information. In the meantime, it used the analog instru- ment cluster to provide information to everyone in the same way. However, nowadays possible to choose the information expression way according to the individual’s taste by using several themes used by the digital instrument cluster [11], [12].

In this paper, we provide a digital instrument cluster ad- justable for the characteristics of each driver. To provide a more effective information providing method, developed a prototype system that provides a digital instrument cluster environment suited to each driver’s characteristic. It is ex- pected that the digital instrument cluster of higher freedom than the existing image resource-based cluster can be used as a base study to derive an effective interaction method by providing an environment suitable for the usefulness evaluates of each driver group. Future research will improve the system to facilitate the replication of the instrument cluster, which is highly similar to that of mass-produced vehicles.

Using the improved system, we will generate the instrument cluster of Korea’s best-selling vehicle this year and use a driving simulator to change the shape, color, and layout of

the instrument cluster’s elements for each group of drivers to evaluate how each driver will cognitive response. To measure the driver’s cognitive response, each driver group differences are derived using button presses, voice feedback, or eye track- ers, and it uses the results to provide a customized instrument cluster according to driver cognitive characteristics.

ACKNOWLEDGMENT This work was supported by Electronics and Telecommuni-

cations Research Institute (ETRI) grant funded by the Korean government and Ulsan Metropolitan City. [19ZS1300, The development of smart context-awareness foundation technique for major industry acceleration], [19AS1100, Development of smart HSE system and digital cockpit system based on ICT convergence for enhanced major industry].

REFERENCES [1] P. M. Knoll, “The use of displays in automotive applications,” J. Soc.

Inf. Display, vol. 5, no. 3, p. 165, 1997. [2] T.-Y. Koo et al., “A study on driver’s workload of telematics using a

driving simulator: A comparison among information modalities,” Int. J. Precis. Eng. Manuf., vol. 10, no. 3, pp. 59–63, Jul. 2009.

[3] S. H. Yoon, J. Lim, and Y. G. Ji, “Assessment model for perceived visual complexity of automotive instrument cluster,” Appl. Ergonom., vol. 46, pp. 76–83, Jan. 2015.

[4] S. H. Yoon, J. H. Lim, and Y. G. Ji, “Perceived visual complexity and visual search performance of automotive instrument cluster: A quantitative measurement study,” Int. J. of Human-Comput. Interact., vol. 31, no. 12, pp. 890–900, Dec. 2015.

[5] F. Schewe and M. Vollrath, “Visualizing distances as a function of speed: Design and evaluation of a distance-speedometer,” Transp. Res. Part F: Traffic Psychol. Behav., vol. 64, pp. 260–273, Jul. 2019.

[6] M. Ilbeygi and M. R. Kangavari, “Comprehensive architecture for intelligent adaptive interface in the field of single-human multiple-robot interaction,” ETRI J., vol. 40, no. 4, pp. 483–498, Aug. 2018.

[7] A. Kuś et al., “Holographic tomography: Hardware and software solu- tions for 3D quantitative biomedical imaging (Invited paper),” ETRI J., vol. 41, no. 1, pp. 61–72, Feb. 2019.

[8] K. Kim et al., “Requirements for older drivers in the driver-adaptive vehicle interaction system,” in Proc. IEEE 8th Int. Conf. Inform. Commun. Technol. Convergence (ICTC), Oct. 2017, pp. 1021–9023.

[9] J.-K. Choi et al., “Conceptual design of driver-adaptive human-machine interface for digital cockpit,” in Proc. IEEE 9th Int. Conf. Inform. Commun. Technol. Convergence (ICTC), Oct. 2018, pp. 1005–1007.

[10] J.-K. Choi et al., “Configurable automotive cluster display considering driver’s cognitive characteristics,” in Proc. IEEE 18th Int. Conf. Electron. Inf. Commun. (ICEIC), Jan. 2019, pp. 804–805.

[11] C. J. Normark, “Design and evaluation of a touch-based personalizable in-vehicle user interface,” Int. J. of Human-Comput. Interact., vol. 31, no. 11, pp. 731–745, Nov. 2015.

[12] P. Kaufmann and A. Riener, “Evaluation of driving performance and user experience of different types of speedometer,” in Proc. ACM 10th Int. Conf. Automot. User Interfaces Interactive Veh. Appl. (AutomotiveUI), 2018, pp. 110–114.

463

Authorized licensed use limited to: Florida Institute of Technology. Downloaded on November 30,2021 at 20:54:02 UTC from IEEE Xplore. Restrictions apply.