HCI Bibliography Home | HCI Conferences | DHM Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
DHM Tables of Contents: 07091113-113-21415-115-2

DHM 2011: 3rd International Conference on Digital Human Modeling

Fullname:DHM 2011: 3rd International Conference on Digital Human Modeling
Note:Volume 17 of HCI International 2011
Editors:Vincent G. Duffy
Location:Orlando, Florida
Dates:2011-Jul-09 to 2011-Jul-14
Publisher:Springer-Verlag
Series:Lecture Notes in Computer Science 6777
Standard No:ISBN: 978-3-642-21798-2 (print), 978-3-642-21799-9 (online); hcibib: DHM11
Papers:58
Pages:532
Links:Online Proceedings | Publisher Book Page
  1. Anthropometry Applications
  2. Posture and Motion Modeling
  3. Digital Human Modeling and Design
  4. Cognitive Modeling
  5. Driver Modeling

Anthropometry Applications

The Effects of Landmarks and Training on 3D Surface Anthropometric Reliability and Hip Joint Center Prediction BIBAKFull-Text 3-11
  Wen-Ko Chiou; Bi-Hui Chen; Wei-Ying Chou
Deforming 3D scanned data is an important and necessary procedure for the development of dynamic three-dimensional (3D) scanned anthropometry. The inaccuracies in joint center will cause error in deformation. Bell et al. developed the equations to predict hip joint center (HJC) based on anthropometric measurement of inter-anterior superior iliac spine distance (IAD). However, no previous study has reported on the reliability of IAD measurements in 3D scanned data, and therefore the effect on HJC estimates needs to be determined. Four measurers (2 trained/ 2 untrained) were recruited into this study to collect measurements of IAD in 3D scanned data under two situations (with/ without landmarks). The intra-class correlation (ICC) and technical error of measurement (TEM) were used to assess the reliability of the measurements. Results showed the untrained group had the lowest reliability and validity of IAD measurement in the without landmarks situation, and the error of HJC prediction in this situation was significantly higher than in the other situations (p<0.001). Both of training and use of landmarks improved the validity of measurement and HJC prediction; compared with training alone, attaching landmarks can significantly improve the reliability of measurement.
Keywords: Reliability; Hip joint center; Three-dimensional body scanner; Trained; Landmarks
An Automatic Method for Computerized Head and Facial Anthropometry BIBAKFull-Text 12-21
  Jing-Jing Fang; Sheng-Yi Fang
Facial anthropometry plays an important role in ergonomic applications. Most ergonomically-designed products depend on stable and accurate human body measurement data. Head and facial anthropometric dimensions provide detailed information on head and facial surfaces to develop well-fitting, comfortable and functionally-effective facial masks, helmets or customized products. Accurate head and facial anthropometry also allows orthognathic surgeons and orthodontists to plan optimal treatments for patients. Our research uses an automatic, geometry-based facial feature extraction method to identify head and facial features, which can be used to develop a highly-accurate feature-based head model. In total, we have automatically located 17 digital length measurements and 5 digital tape measurements on the head and face. Compared to manual length-measurement, the average error, maximum error and standard deviations are 1.70mm, 5.63mm and 1.47mm, respectively, for intra-measurement, and 2.07mm, 5.63mm and 1.44mm, respectively, for inter-measurement. Compared to manual tape-measurement, the average maximum error and standard deviations are 1.52mm, 3.00mm and 0.96mm, respectively, for intra-measurement, and 2.74mm, 5.30mm and 1.79mm, respectively, for inter-measurement. Nearly all of length measurement data and tape measurement data meet the 5mm measuring error standard.
Keywords: anthropometry; head and face; feature-based
3D Parametric Body Model Based on Chinese Female Anthropometric Analysis BIBAKFull-Text 22-29
  Peng Sixiang; Chee Kooi Chan; W. H. Ip; Ameersing Luximon
This study shows a 3D parametric body model construction based on the anthropometric analysis technique. Compared to the traditional anthropometric surveys, the 3D body scanner provides with more accurate body dimension information and not only traditional measurements but also new body shape measurements. An anthropometric survey was completed to collect 3D body information of Hong Kong female. A serious of 3D models was built upon these 3D scan data. A number of body dimensions and body shape information were extracted from the 3D models. Then body shape analysis were performed, such as bust shape analysis, front and back proportion analysis, body cross section comparison, correlation relationship between body dimensions and so on. Upon the analysis of the body information, a parametric body model was built, which was in a most common body shape of the Hong Kong female and able to change critical body demotions according to the user's inputs.
Keywords: 3D; parametric; body model
Anthropometric Measurement of the Feet of Chinese Children BIBAKFull-Text 30-36
  Linghua Ran; Xin Zhang; Chuzhi Chao; Taijie Liu
This paper presents the results of a nationwide anthropometric survey conducted on children in China. Foot length and foot breadth were measured from 20,000 children with age ranged from 4 to 17 years old by 3D foot scanner. Mean values, standard deviations, and the 5th, 95th percentile for the two items were estimated. The dimension difference between age and gender were discussed, and the classification of foot shape was analyzed. It was found that the mean values of the dimensions showed a gradual increase by age. The dimensions had no significant difference between genders for the children from 4 to 12, but the difference became significant for the children from 13 to 17. For both boys and girls, the intermediate type foot has the greatest proportion. These data, previously lacking in China, can benefit the children's products design.
Keywords: Foot; anthropometric measurement; Chinese children
Human Dimensions of Chinese Minors BIBAKFull-Text 37-45
  Xin Zhang; Yanyu Wang; Linghua Ran; Ailan Feng; Ketai He; Taijie Liu; Jianwei Niu
This paper presents the preliminary statistical results of the latest national anthropometric survey of the minors in China mainland. About 20,000 minors (9666 males and 9699 females) were recruited from six geographical areas in China. These subjects were divided into five age groups. Body weight plus totally 134 static dimensions were selected for measurement. Non-contact three-dimensional (3D) scanning technology was used in this survey while manual measuring and two-dimensional (2D) imaging measurement were used as the subsidiary methods. There is no significant difference on body height between genders at the significance level of 0.05 when the subjects are less than 12 years old. The 5th, 50th and 95th percentile values for nineteen selected dimensions were addressed. This work provides the first national wide anthropometric database of the minors in China mainland and will inevitable benefit the products design for the potential users.
Keywords: Anthropometric; minors; three dimension; survey
Development of Sizing Systems for Chinese Minors BIBAKFull-Text 46-55
  Xin Zhang; Yanyu Wang; Linghua Ran; Ailan Feng; Ketai He; Taijie Liu; Jianwei Niu
The purpose of this study is to develop sizing systems for Chinese minors. This work is based on the most up-to-date and complete national-wide anthropometric survey in China mainland. About 20,000 minors (9666 males and 9699 females) were recruited from six geographical areas in China. Body weight plus totally 134 static dimensions were collected. Stature, the bust waist girth difference (BWGD) and bust girth were finally identified through factor analysis as the three most critical parameters out of thirteen frequently used anthropometric dimensions in sizing systems. Three sizing systems were established systematically by stature group and gender, i.e., both genders shorter than 130 cm, male minors taller than 130 cm and female minors taller than 130 cm. The accommodation rate of the developed sizing systems is 47.23%, 60.19% and 81.87%, respectively. The number of sizes in each system is 25, 25 and 19, respectively. The results provide valuable references for the garment manikin design related with Chinese minors.
Keywords: Anthropometric data; Sizing system; Chinese minors; Factor analysis

Posture and Motion Modeling

Motion Capture Experiments for Validating Optimization-Based Human Models BIBAKFull-Text 59-68
  Aimee Cloutier; Robyn Boothby; Jingzhou (James) Yang
Optimization-based digital human model research has gained significant momentum among various human models. Any task can be formulated to an optimization problem, and the model can predict not only postures but also motions. However, these optimization-based digital human models need validation using experiments. The motion capture system is one of the ways to validate predicted results. This paper summarizes the progress of motion capture experiment efforts at the Human-Centric Design Research (HCDR) Laboratory at Texas Tech University. An eight-camera motion capture system has been set up in our research lab. Marker placement protocols have been developed where markers are placed on the subjects to highlight bony landmarks and identify segments between joints in line with previously identified guidelines and suggestions in literature. A posture reconstruction algorithm has been developed to map joint angles from motion capture experiments to digital human models. Various studies have been conducted in the lab involving motion capture experiments for jumping, standing and seated reach, and pregnant women's walking, sit to standing, seated reach, and reach with external loads. The results showed that the posture reconstruction algorithm is useful and accurate to transfer motion capture experiment data to joint angles. Marker placement protocol is reliable to capture all joints. The main task of the motion caption system is to validate all optimization-based digital human models developed by other research members at the HCDR Lab.
Keywords: Digital human models; motion capture; validation
Posture Reconstruction Method for Mapping Joint Angles of Motion Capture Experiments to Simulation Models BIBAKFull-Text 69-78
  Jared Gragg; Jingzhou (James) Yang; Robyn Boothby
Motion capture experiments are often used in coordination with digital human modeling to offer insight into the simulation of real-world tasks or as a means of validating existing simulations. However, there is a gap between the motion capture experiments and the simulation models, because the motion capture system is based on Cartesian space while the simulation models are based on joint space. This paper bridges the gap and presents a methodology that enables one to map joint angles of motion capture experiments to simulation models in order to obtain the same posture. The posture reconstruction method is an optimization-based approach where the cost function is a constant and constraints include (1) the distances between simulation model joint centers and the corresponding experimental subject joint centers are equal to zeros; (2) all joint angles are within joint limits. Examples are used to demonstrate the effectiveness of the proposed method.
Keywords: digital human modeling; posture reconstruction; motion capture
Joint Torque Modeling of Knee Extension and Flexion BIBAKFull-Text 79-88
  Fabian Guenzkofer; Florian Engstler; Heiner Bubb; Klaus Bengler
The purpose of this experiment is to obtain isometric knee extension and flexion joint torque -- joint angle functions considering necessary biomechanical aspects. In order to examine gender and age effects four different subject groups (10 young males and females, 8 old males and females) were used. Age and gender had a significant influence for both force directions. Not only different maximum values but also different curve shapes were identified for different age groups. Additionally the hip flexion angle significantly influenced the joint torque production.
Keywords: joint torque; age effects; strength; knee; force
Predicting Support Reaction Forces for Standing and Seated Tasks with Given Postures-A Preliminary Study BIBAKFull-Text 89-98
  Brad Howard; Jingzhou (James) Yang
This paper proposes a systematic approach for predicting the support reaction forces (SRFs) acting on a digital human model with a given posture. In addition, a generic method has been developed to determine the accurate body segment inertia properties (BSIPs) needed for subject-specific simulation. Experiments based on motion capture are used to track the posture and to find subject's link lengths. The prediction model calculates the support reaction forces by using the zero moment point (ZMP) formulation. This study considers two general postural cases: standing and seated. The standing tasks include standing on two planes with arbitrary orientations. The seated tasks include sitting on a seat where the seat pan is parallel to the floor and both feet are on the floor.
Keywords: Support reaction forces; digital human model; ZMP; posture
Schema for Motion Capture Data Management BIBAKFull-Text 99-108
  Ali Keyvani; Henrik Johansson; Mikael Ericsson; Dan Lämkull; Roland Örtengren
A unified database platform capable of storing both motion captured data and information about these motions (metadata) is described. The platform stores large motion captured data in order to be used by different applications for searching, comparing, analyzing and updating existing motions. The platform is intended to be used to choose a realistic motion in simulation of production lines. It is capable of supporting and handling different motion formats, various skeleton types and distinctive body regions in a uniform data model. Extended annotating system is also introduced to mark the captured data not only in the time domain (temporal) but also on different body regions (spatial). To utilize the platform, sample tests are performed to prove the functionality. Several motion captured data is uploaded to the database while MATLAB is used to access the data, ergonomically analyze the motions based on OWAS standard, and add the results to the database by automatic tagging of the postures.
Keywords: Motion Capture Database; Virtual Production Systems; Digital Human Modeling; Computerized Ergonomic Analysis
Simulating Ingress Motion for Heavy Earthmoving Equipment BIBAKFull-Text 109-118
  HyunJung Kwon; Mahdiar Hariri; Rajan Bhatt; Jasbir Arora; Karim Abdel-Malek
Design of heavy earth moving equipment is based primarily on feedback from the driver. Most design studies on ingress focus on the motion itself and rely heavily on experimental data. This process requires physical construction of an expensive mockup before any feedback can be obtained. Moreover, most research and development on subject of ingress are limited to studies on passenger vehicles. Although the design of heavy vehicles requires more consideration to human safety and comfort, very little attention has been given to simulating ingress movement on those vehicles. This paper describes the development of a model to perform ingress motion for heavy equipment and its applications to study the response of the operator for different cab designs using Santos™, the digital human model developed at Virtual Soldier Research at the University of Iowa.
Keywords: Human Modeling; Ingress; Predictive Dynamics; Heavy Equipment
Contact Area Determination between a N95 Filtering Facepiece Respirator and a Headform BIBAKFull-Text 119-128
  Zhipeng Lei; Jingzhou (James) Yang
This study investigates two methods to determine the contact area between a N95 filtering facepiece (NFF) respirator and a headform. Five size headforms (large, medium, small, long/narrow and short/wide) wear a NFF respirator (3M 8210) respectively. A biofidelity finite element model of headform is built to simulate its interaction with a NFF respirator. During the simulation, the respirator contacts the headform. Two methods are presented in this paper for determining contact areas: The first one is through the observation of contact pressure distribution. The assumption is that the contact area is the fraction of surface area with positive contact pressure. The second method is through extracting the intersecting area between deformed surfaces of the headform and NFF respirator. Finally, the experiment, which directly measures the dimensions of contact area between prototypes of the headform and the NFF respirator, validates the proposed methods.
Keywords: Headform; respirator; finite element (FE) method; contact area
Ergonomics Evaluation of Three Operation Postures for Astronauts BIBAKFull-Text 129-138
  Dongxu Li; Yan Zhao
Push/pull is very common and frequent activity for astronauts in the space operation. For the sake of researching on the strength change of astronauts' upper limbs during different operation postures, this paper major on performed the simulation aviation operation experiment. The paper quantitatively evaluates the absolute peak force, relative peak force of human single left/right hand and strain energy of grip made by operating force in comfortable, horizontal widest, and longitudinal widest span. The result here demonstrates that absolute peak force, relative peak force, strain energy are effective factors to estimate upper limbs strength, and the individual upper limbs power is significantly effected by different postures.
Keywords: astronaut; upper limb; absolute peak force; relative peak force; strain energy
In Silicon Study of 3D Elbow Kinematics BIBAKFull-Text 139-142
  Kang Li; Virak Tan
This study is to propose a novel technique to improve the accuracy of estimating bone kinematics. This technique will use the radiographic information of both soft tissue and hard tissue for the 2D-3D registration. Non-rigid registration technique and rigid-body registration will work seamlessly to guide the matching process to find the optimal bone pose. Such a technique could improve and accelerate the matching process.
Keywords: Elbow; Kinematics; Imaging
Implicit Human-Computer Interaction by Posture Recognition BIBAKFull-Text 143-150
  Enrico Maier
The presented work introduces the evolution of the computer to an ubiquitous attendant of the human. The related incessantly opportunity to interact with the computer and the explicit character of current user interfaces leads to the distraction of the human from his actual tasks. Accordingly, the authors of this work suggest the design and the research of new and implicit interaction forms, which support the human while the processing of his tasks. In this work, a concept for the implicit human-computer interaction by the recognition of prejudicial postures will presented. The detection and feedback of prejudicial postures by mobile computers like smartphones shall warn the human of unhealthy postures in everyday life and on work and prevent him from sanitary after effects.
Keywords: human-computer interaction; implicit interaction; mobile interaction; ubiquitous computing
Optimization-Based Posture Prediction for Analysis of Box Lifting Tasks BIBAKFull-Text 151-160
  Tim Marler; Lindsey Knake; Ross Johnson
New methods for optimization-based posture prediction with external forces are presented and tested. The proposed approach incorporates prediction of 113 degrees of freedom including global position and orientation of the body as well as foot position, while considering balance. Postures and joint torques are successfully predicted and compared to motion-capture data and literature-based data respectively. This approach is applied to a box-lifting task and provides a robust tool for studying human performance and for preventing injuries.
Keywords: Posture prediction; optimization; box lifting; joint torque
Planar Vertical Jumping Simulation-A Pilot Study BIBAKFull-Text 161-170
  Burak Ozsoy; Jingzhou (James) Yang
Vertical jumping is one of the fundamental motions among other jumping types in sport biomechanics. Two important criteria in sport biomechanics are critical to all athletes: Injury and performance. In literature two major approaches have been investigated: experiment-based methods and optimization-based methods. Experiment-based methods are time consuming and tedious. Optimization-based methods for musculoskeletal models are computationally expensive because their models include all muscles and explicit integration of equation of motion. In this pilot study, a direct optimization-based method for a skeletal model was proposed in sagittal plane, where this formulation was based on joint space that was only considered the resultant results of muscles (joint torques) instead of individual muscles to reduce computational time. The cost function included increasing the center of mass velocity at take-off and increasing the center of mass position at take-off. Constraints included joint limits, torque limits, initial posture, ground contact, initial angular velocity and acceleration, zero-ground reaction forces, and moment at take-off. This optimization problem was solved by a commercial optimization solver SNOPT and the CPU time was 227 seconds on a regular PC (Intel® Core® 2 duo CPU, 3.16 GHZ and 3.25 GB RAM). Preliminary results highly correlated results from the literature. This simple planar simulation is the first step to understand the cause and effect for vertical jumping with or without arm swing.
Keywords: Vertical jumping; planar model; injury; performance; arm swing
StabilitySole: Embedded Sensor Insole for Balance and Gait Monitoring BIBAKFull-Text 171-177
  Peyton Paulick; Hamid Djalilian; Mark Bachman
Our group has developed an easy-to-use pressure-sensing insole equipped with accelerometers to evaluate a patient's postural control and balance ability in a non-obtrusive manner. Pressure sensors and MEMS accelerometers are embedded in a shoe insole and display balance information on an easy-to-use graphical user interface. The importance of balance and gait monitoring is applicable to a variety of fields such as: evaluation of neurological disorders, early disease detection in children, new medication monitoring, and elderly fall risk. With this technology patients can use the insole in a home setting to monitor their balancing ability and relay data to their physician wirelessly for proper evaluation eliminating the need for additional in-office visits.
Keywords: Balance monitoring; gait monitoring; insole; embedded technology; graphical user interface; pressure sensing; MEMS accelerometer
The Upper Extremity Loading during Typing Using One, Two and Three Fingers BIBAKFull-Text 178-185
  Jin Qin; Matthieu Trudeau; Jack Tigh Dennerlein
This study aimed to evaluate the effect of the number of fingers used during typing on the biomechanical loading on the upper extremity. Six subjects typed in phone numbers using their right hand on a stand-alone numeric keypad in three conditions: (1) typing using the index finger; (2) typing using the index and the middle fingers; (3) typing using the index, middle and ring fingers. Typing with three fingers decreased wrist posture deviation, decreased angular velocity at the wrist, elbow and shoulder joints, and decreased peak to peak torques at the wrist and shoulder joints compared to single finger typing, while no difference was found between one and two finger typing. These results demonstrated that different computer keyboarding styles affect the biomechanical loading on the upper extremity.
Keywords: Typing; Upper extremity; Kinematics; Kinetics
Automatic Face Feature Points Extraction BIBAKFull-Text 186-194
  Dominik Rupprecht; Sebastian Hesse; Rainer Blum
In this paper we present results of finding a way to automatically equip a three-dimensional avatar with a model of a user's individual head. For the generation of the head model certain so-called face feature points must be extracted from a face picture of the user. A survey of several state-of-the-art techniques and the results of an approach for the extraction process are given for the points of the middle of each iris, the nasal wings and the mouth corners.
Keywords: Avatar; Face Feature Points; Integral Projection; Circle Detection; E-Commerce
3D Human Motion Capturing Based Only on Acceleration and Angular Rate Measurement for Low Extremities BIBAKFull-Text 195-203
  Christoph Schiefer; Thomas Kraus; Elke Ochsmann; Ingo Hermanns; Rolf P. Ellegast
Human motion capturing is used in ergonomics for ambulatory assessment of physical workloads in field. This is necessary to investigate the risk of work-related musculoskeletal disorders. Since more than fifteen years the IFA is developing and using the motion and force capture system CUELA, which is designed for whole-shift recordings and analysis of work-related postural and mechanical loads. A modified CUELA system was developed based on 3D inertial measurement units to replace all mechanical components in the present system. The unit consists of accelerometer and gyroscopes, measuring acceleration and angular rate in three dimensions. The orientation determination based on angular rate has the risk of integration errors due to sensor drift of the gyroscope. We introduced "zero points" to compensate gyroscope drift and reinitialize orientation computation. In a first evaluation step the movements of lower extremities are analyzed and compared to the optical motion tracking system Vicon.
Keywords: ambulatory workload assessment; inertial tracking device; motion capturing; CUELA; ergonomic field analysis
Application of Human Modeling in Multi-crew Cockpit Design BIBAKFull-Text 204-209
  Xiaohui Sun; Feng Gao; Xiugan Yuan; Jingquan Zhao
Based on the need of multi-crew cockpit ergonomic design, we set up a parameterized digital human model. By virtually controlling digital human model to devices of cockpit, the potential conflicts in the process of multi-crew coordination were identified. The solutions on device layout were proposed. This method beforehand took human factors into account in the multi-crew cockpit design. Design efficiency was improved greatly.
Keywords: Human modeling; Cockpit design; Crew Coordination
A Biomechanical Approach for Evaluating Motion Related Discomfort: Illustration by an Application to Pedal Clutching Movement BIBAKFull-Text 210-219
  Xuguang Wang; Romain Pannetier; Nagananda Krishna Burra; Julien Numa
In this paper, a motion related discomfort modelling approach based on the concept of "less constraint movement" has been proposed and illustrated by a case study of clutching pedal movements. Using a multi-adjustable car mock-up, 6 existing pedal configurations were tested by 20 subjects (5 young and 5 older males, 5 young and older females) and compared with those freely adjusted pedal positions, called 'less constraint' configurations. From questionnaire and motion analysis of the experimental data, it was observed that pedal resistance had a dominant effect on discomfort perception. The pedal position adjustment seemed to mainly reduce the discomfort at the beginning of travel. The ergonomic criterion for pedal design should therefore take into account two main factors: 1/ pedal resistance, 2/ a good trade-off between pedal position at the beginning of travel and its end position. The movements corresponding to less constraint configurations will be used as reference data for defining ergonomic criterion. In particular, The focus should be put on the hip, knee and ankle joint torques at the end of travel and the joint angles at the beginning and end of pedal travel.
Keywords: Discomfort; Clutch pedal; Biomechanics; Ergonomics; Task-oriented movement
Footbed Influences on Posture and Perceived Feel BIBAKFull-Text 220-227
  Thilina W. Weerasinghe; Ravindra S. Goonetilleke
Past studies have evaluated body postures when wearing high-heeled shoes. However, the effects of the various parameters that give the footbed its shape have not been investigated. This study determined the perceived feel and the associated postures with the different types of footbed shapes at two heel heights of 50 and 75 mm. Results show that a 10 degree wedge angle at 50 mm and an 18 degree wedge angle at 75 mm have the highest perceived feel during standing. The corresponding postures were significantly different from the others tested, suggesting that a well-designed footbed is a necessary condition to feel comfortable and to maintain good posture.
Keywords: Posture; Footbed; Comfort; High-heels; Footwear
Postural Observation of Shoulder Flexion during Asymmetric Lifting Tasks BIBAKFull-Text 228-230
  Xu Xu; Chien-Chi Chang; Gert S. Faber; Idsart Kingma; Jack Tigh Dennerlein
This study was to evaluate the observation error of the shoulder flexion angle during an asymmetric lifting task. The results indicated the average absolute estimate error was 14.7 degrees and the correlation coefficient between the measured and estimated shoulder flexion was 0.91. The observation error may be due to the arm abduction.
Keywords: Shoulder flexion; Posture observation; Side-view; Lifting tasks
An Alternative Formulation for Determining Weights of Joint Displacement Objective Function in Seated Posture Prediction BIBAKFull-Text 231-242
  Qiuling Zou; Qinghong Zhang; Jingzhou (James) Yang; Robyn Boothby; Jared Gragg; Aimee Cloutier
The human posture prediction model is one of the most important and fundamental components in digital human models. The direct optimization-based method has recently gained more attention due to its ability to give greater insights, compared to other approaches, as how and why humans assume a certain pose. However, one longstanding problem of this method is how to determine the cost function weights in the optimization formulation. This paper presents an alternative formulation based on our previous inverse optimization approach. The cost function contains two components. The first is the weighted summation of the difference between experimental joint angles and neutral posture, and the second is the weighted summation of the difference between predicted joint angles and the neutral posture. The final objective function is then the difference of these two components. Constraints include (1) normalized weights within limits; (2) an inner optimization problem to solve for the joint angles, where joint displacement is the objective function; (3) the end-effector reaches the target point; and (4) the joint angles are within their limits. Furthermore, weight limits and linear weight constraints determined through observation are implemented. A 24 degree of freedom (DOF) human upper body model is used to study the formulation. An in-house motion capture system is used to obtain the realistic posture. Four different percentiles of subjects are selected and a total of 18 target points are designed for this experiment. The results show that using the new objective function in this alternative formulation can greatly improve the accuracy of the predicted posture.
Keywords: Posture prediction; direct optimization-based posture prediction; digital human

Digital Human Modeling and Design

Videogames and Elders: A New Path in LCT? BIBAKFull-Text 245-254
  Nicola D'Aquaro; Dario Maggiorini; Giacomo Mancuso; Laura Anna Ripamonti
The current demographic ageing in Europe is the result of a relevant economic, social, and medical development. Nevertheless, this is also leading to an increase in the demand for Long Term Care (LTC) by seniors. One viable way to offer qualified cares at home, while at the same time containing costs, is to exploit digital technologies as enablers of a constant interaction with assisting personnel. The main contribution of this paper is to propose a design methodology to put the basis for deploying, step by step, a "virtual hospital at home" based on an Ambient Intelligence (AmI). The envisaged system will integrate consumer-grade devices for videogames consoles which, thanks to their user-friendly interfaces and smooth learning curve, will contribute in minimizing the interference in the elder's private life.
Keywords: Ambient Intelligence (AmI); context awareness; videogames; healthcare; LTC (Long Term Care); usability
Research on Digital Human Model Used in Human Factor Simulation and Evaluation of Load Carriage Equipment BIBAKFull-Text 255-262
  Dayong Dong; Lijing Wang; Xiugan Yuan; Shan Fu
Data structure of the digital human model has been constructed for particularity demand of human factor evaluation of the load carriage system equipments. Anthropometry data of different percentage was obtained through regression calculation; Range of Motion (ROM) data of the different joints, data of upper limb reach zone and accessibility data of human body surface were also obtained by experiments. Reach zone envelop was constructed based on the experiment data. Different types of view cones have been constructed on the basis of related standard data to meet the use of human factor evaluation.
Keywords: Load carriage equipment; Simulation; Human factors; Digital Human model
Multimodal, Touchless Interaction in Spatial Augmented Reality Environments BIBAKFull-Text 263-271
  Monika Elepfandt; Marcelina Sünderhauf
Spatial augmented reality environments are increasing in many domains. However, from a user-centred perspective we still do not know how to interact with them properly. Thus, there is a need for new interaction concepts. In this paper we discuss certain aspects which are special for the interaction in spatial augmented reality environments. We conclude that the most promising form of interaction will be multimodal and touchless. Additionally it should be taken into account that the user is interacting in a 3D space. Therefore different ways of interaction might be more appropriate for different spaces around the user instead of only one. We give a short review of advantages and disadvantages of touchless and multimodal input devices and then outline spatial aspects of perception and interaction in 3D space.
Keywords: spatial augmented reality; touchless interaction; multimodal interaction; multimodality; interaction in 3D space
Introducing ema (Editor for Manual Work Activities) -- A New Tool for Enhancing Accuracy and Efficiency of Human Simulations in Digital Production Planning BIBAKFull-Text 272-281
  Lars Fritzsche; Ricardo Jendrusch; Wolfgang Leidholdt; Sebastian Bauer; Thomas Jäckel; Attila Pirger
The aging workforce is a risk factor for manufacturing industries that contain many jobs with high physical workloads. Thus, ergonomic risk factors have to be avoided in early phases of production planning. This paper introduces a new tool for simulating manual work activities with 3D human models, the so-called ema. For the most part, the ema software is based on a unique modular approach including a number of complex operations that were theoretically developed and empirically validated by means of motion capturing technologies. Using these modules for defining the digital work process enables the production planner to compile human simulations more accurately and much quicker compared to any of the existing modeling tools. Features of the ema. software implementation, such as ergonomic evaluation and MTM-time analyses, and the workflow for practical application are presented.
Keywords: Human Modeling; Production Planning; Ergonomics; Efficiency
Accelerated Real-Time Reconstruction of 3D Deformable Objects from Multi-view Video Channels BIBAKFull-Text 282-291
  Holger Graf; Leon Hazke; Svenja Kahn; Cornelius Malerczyk
In this paper we present a new framework for an accelerated 3D reconstruction of deformable objects within a multi-view setup. It is based on a new memory management and an enhanced algorithm pipeline of the well known Image-Based Visual Hull (IBVH) algorithm that enables efficient and fast reconstruction results and opens up new perspectives for the scalability of time consuming computations within larger camera environments. As a result, a significant increase of frame rates for the volumetric reconstruction of deformable objects can be achieved using an optimized CUDA-based implementation on NVIDIA's Fermi-GPUs.
Keywords: Image based 3D reconstruction; GPU; CUDA; Real-Time reconstruction; Image-Based Visual Hull
Second Life as a Platform for Creating Intelligent Virtual Agents BIBAKFull-Text 292-301
  Larry F. Hodges; Amy Catherine Ulinski; Toni Bloodworth; Austen L. Hayes; John Mark Smotherman; Brandon Kerr
Intelligent virtual agents (IVAs) are animated characters that interact with humans and with each other using natural modalities such as speech and gestures. Creation of successful IVA applications is challenging since development requires integration of a broad range of specialized tools and skills. We present a discussion of the advantages and technical challenges of using the Second Life programming environment as an accessible platform for undergraduates to develop an Intelligent Virtual Agent and the environment that the IVA inhibits.
Keywords: Intelligent Virtual Agents; Virtual Worlds; Animation; Avatars; Virtual Characters
A Framework for Automatic Simulated Accessibility Assessment in Virtual Environments BIBAKFull-Text 302-311
  Nikolaos Kaklanis; Panagiotis Moschonas; Konstantinos Moustakas; Dimitrios Tzovaras
The present paper introduces a framework that enforces the accessibility of products and services by enabling automatic simulated accessibility assessment at all the stages of the development. The proposed framework is based on a new virtual user modelling technique describing in detail all the physical parameters of a user with disability(ies). The proposed user modelling methodology generates a dynamic and parameterizable virtual user model that is used by a simulation framework to assess the accessibility of virtual prototypes. Experimental results illustrate the use of the proposed framework in a realistic application scenario.
Keywords: accessibility evaluation; user modelling; task modelling; simulation; UsiXML; virtual user
Cloth Modeling and Simulation: A Literature Survey BIBAKFull-Text 312-320
  James Long; Katherine Burns; Jingzhou (James) Yang
Cloth modeling and simulation has gained significant momentum in recent years due to advances in computational hardware and software. Cloth plays an important role not only in daily life, but also in special scenarios such as firefighter's cloth and space suits. There are special requirements such as protection capability of the human body and mobility after the firefighters or astronauts wear the special cloth. Traditional assessment of cloth is to have prototypes first and have experiments by subjective rating. This is time consuming and expensive. Virtual cloth modeling and simulation provides a means to demonstrate and assess its performance before cloth is made. This paper attempts to give a literature review to summarize the state-of-the-art of cloth modeling and simulation.
Keywords: Cloth modeling and simulation; finite element method; comfort
Preliminary Study on Dynamic Foot Model BIBAFull-Text 321-327
  Ameersing Luximon; Yan Luximon
It is generally accepted that improper footwear design causes injuries and illnesses. Existing literature indicated that mostly static footwear fit has been studied to some extent, even though we are well aware that dynamic footwear fit is different. In addition to static fit, illnesses and injuries will occur due to dynamic pressure, friction, and foot movement beyond the normal range of motion. This study proposes a method for dynamic foot shape computation. The dynamic foot is generated by using multi-dimensional transformation of the 3D static foot based on the angular values of the foot movement. This provides a simple algorithm to deform the foot to create dynamic 3D foot shape. Result of this study is essential to build a model for dynamic fit computation.
Three-Dimensional Grading of Virtual Garment with Design Signature Curves BIBAKFull-Text 328-336
  Roger Ng
The key difference between tailoring and mass production is the use of a single size versus a size chart that contains difference sizes. In practice, the garment pattern of a reference size is made for the confirmation of style and sizing during the product development process. Then, garment patterns of other sizes are derived by this reference size using a technique of grading. In the flat-patterning techniques, there are three types of grading: rectangular (Cartesian coordinate), radio (polar coordinate) and line (localized Cartesian coordinate). All these methods suffer from the limitation of the increase of deformation as the sizes increase. In this article, I shall present a three-dimensional method of grading, which can maintain the styling and comfort characteristics of the wearers at different sizes. This is achieved by the concept of Design Signature Curve. After I present the concept, I shall also present an example as the verification.
Keywords: 3-D Grading; Design Signature Curve; Design; Fitting; Apparel
A Model of Shortcut Usage in Multimodal Human-Computer Interaction BIBAKFull-Text 337-346
  Stefan Schaffer; Robert Schleicher; Sebastian Möller
Users of multimodal systems have to choose between different interaction strategies. Thereby the number of interaction steps to solve a task can vary across the available modalities. In this work we introduce such a task and present empirical data that shows that strategy selection of users is affected by modality specific shortcuts. The system under investigation offered touch screen and speech as input modalities. We introduce a first version of an ACT-R model that uses the architectures-inherent mechanisms production compilation and utility learning to identify modality-specific shortcuts. A simple task analysis is implemented in declarative memory. The model reasonably accurate matches the human data. In our further work we will try to get a better fit by extending the model with further influence factors of modality selection like speech recognition errors. Further the model will be refined concerning the cognitive processes of speech production and touch screen interaction.
Keywords: Multimodal HCI; User Modeling; Automated Usability Evaluation
Multimodal User Interfaces in IPS² BIBAKFull-Text 347-356
  Ulrike Schmuntzsch; Matthias Rötting
The field of Industrial Product-Service Systems (IPS2) faces various challenges. Goal of the recently started project is the multimodal design of interaction specific warnings and instructions in IPS2 for human operators with relation to their interactions. This approach should help to prevent mistakes of human operators whilst interaction and it should account for the support and optimization of work process in the heterogeneous area of IPS2. Here we discuss how such a project can be realized step-by-step.
Keywords: Multimodal user interfaces; interaction specific warnings; user generated instructions; IPS2
The Application of the Human Model in the Thermal Comfort Assessment of Fighter Plane's Cockpit BIBAKFull-Text 357-366
  Haifeng Shen; Xiugan Yuan
Thermal comfort is an important content concerned by aircraft designing and using department. Computational Fluid Dynamics (CFD) is a main numerical simulation method used in this field. A human model of a pilot is built according to its geometrical and thermal features in order to simulate the air flow and heat transfer in a fighter plane's cockpit. The velocity and temperature fields are obtained by using a numerical simulation flat involving the human model above. Meanwhile, an experiment was carried out to prove the effectiveness of numerical simulation. The comparison between calculation results and experiment results shows that the calculation error of velocity and temperature is 15% and 6% respectively, which proves that the human model can be used for numerical simulation and satisfy the requirement of simulation precision. Furthermore, some thermal comfort assessment criteria such as the mean skin temperature and the core temperature of the body are obtained by putting calculation result above into the Human Thermal Regulation System (HTRS) model.
Keywords: cockpit; CFD; human model; thermal comfort; assessment
Mass Customization Methodology for Footwear Design BIBAKFull-Text 367-375
  Yifan Zhang; Ameersing Luximon; Xiao Ma; Xiaoling Guo; Ming Zhang
Nowadays consumers are sophisticated and want better quality personalized products at lower prices. In order to satisfy consumers, footwear companies have to predict consumers' trends and produce huge varieties of designs using mass production technologies. This results in wastage and high inventory cost. Recently mass customization methodologies have been proposed. Mass customization enables the design and production of an almost custom made product at almost mass production costs. This can be achieved by reducing the wastage and inventory cost. The products are designed specifically for the consumer, eventually improving consumer satisfaction and the companies' market share. Many researchers have discussed the framework of mass customization; however, there is limited study on the system details, especially for footwear design. The main purpose of this research is to propose a methodology for the mass customization of footwear -- both style and fit customization.
Keywords: Footwear design; Mass customization; Custom footwear; Style customization; Fit customization; Computer-Aided design (CAD) technology

Cognitive Modeling

Incorporating Motion Data and Cognitive Models in IPS² BIBAKFull-Text 379-386
  Michael Beckmann; Jeronimo Dzaack
In the SFB/TR29 a focus lies on human factors and their integration into Industrial Product-Service Systems (IPS2). These innovative systems are complex and dynamic. Human operators need to be able to perform a multitude of complex tasks in such socio-technical systems, providing a challenge to the operators because of the high complexity. Therefore automatic assistance systems are necessary for the overall reliability and effectiveness of such a system. This article describes a theoretical approach for simulating human behavior with cognitive models. The performed actions are recognized with motion capturing in combination with machine learning. By evaluating the perceived action and reality a description for the situation can be automatically generated in real time. This can be used for e.g. providing the human operator with real time contextual feedback.
Keywords: Cognitive User Models; Error Prevention; Human-Machine Interaction; Machine Learning; Gesture Recognition
Study on Synthetic Evaluation of Human Performance in Manually Controlled Spacecraft Rendezvous and Docking Tasks BIBAKFull-Text 387-393
  Ting Jiang; Chunhui Wang; Zhiqiang Tian; Yongzhong Xu; Zheng Wang
Manual-Control Rendezvous and Docking (MCRVD) is a complex and demanding task which requires astronauts to control the direction of six-freedom of the spacecraft accurately by hand. Its performance has a close relationship among the design of the spacecraft HCI, the control ability of the astronaut, and the matching effect of the two factors above. In this paper, human performance for these tasks was measured and a mathematical model to evaluate MCRVD performance quantitatively was proposed. First, 3500 experiments were designed on a ground simulated RVD system to examine characteristics and regulations of MCRVD performance indexes, such as control deviation, fuel consumption etc. Twenty-five male volunteers aged 25-35 participated in the experiment. Analysis predicts that the performance indexes of MCRVD show the characteristics and laws of stages. The process of MCRVD can be divided into three stages: tracking control (about more than 20m distance), accurate control (less than 20m distance) and docking stage (0m). The performance indexes of tracking and accurate control show the relevance of characteristics, and the precision index of docking reflects the difficulty of control the direction of the spacecraft. For that reason, several statistics techniques, such as the factor analysis method, the entropy analysis method etc, are utilized to analyze the weight coefficient of each performance index. After that we presented a novel multi-hierarchy integrated evaluation method, which includes four hierarchies: the performance of the tracking control, the performance of the accurate control, the precision of docking and the result of docking. Finally we utilized this method to analyze the human performance in MCRVD, which verifies the validity of our method.
Keywords: Manual-control Rendezvous and Docking; Human Performance; evaluation method
Dynamic Power Tool Operation Model: Experienced Users vs. Novice Users BIBAFull-Text 394-398
  Jia-Hua Lin; Raymond W. McGorry; Chien-Chi Chang
Previous study demonstrated that a single-degree-of-freedom mechanical model can represent a human power tool operator subjective to impulsive torque reactions. Using only novice tool users, it was shown that the mechanical capabilities to respond to tool torque reaction depended on workstation location and orientation, and varied among users (Lin, Radwin, & Richard, 2000). It was hypothesized that the mechanical model elements among experienced tool operators may be different from novice users. A laboratory study was carried out to measure the equivalent mechanical parameters among the novice and experienced tools users. The results demonstrate the difference between the two groups. Those may represent the strategy developed by the experienced users to minimize the impacts from the impulsive torque reactions.
An Empirical Study of Disassembling Using an Augmented Vision System BIBAKFull-Text 399-408
  Barbara Odenthal; Marcel Ph. Mayer; Wolfgang Kabuß; Bernhard Kausch; Christopher M. Schlick
Within the Cluster of Excellence "Integrative Production Technology for High-Wage Countries" of RWTH Aachen University a numerical control unit and its ergonomic human-machine interface are developed for a robotized production unit. In order to cope with novel systems, the human operator will have to meet new challenges regarding the work requirements. Therefore, based on a first prototype of an augmented vision system (AVS) assisting the human operator in error detection the system is enhanced in order to deal with the task of error correction. Laboratory tests have been performed to find a preferable solution to support the human operator in disassembling of the erroneous object in order to correct the detected error in cooperation with the robot. Significant effects of the supporting mode regarding human assembly time will be presented.
Keywords: Augmented Reality; Disassembly; Cognitive Automation
Polymorphic Cumulative Learning in Integrated Cognitive Architectures for Analysis of Pilot-Aircraft Dynamic Environment BIBAFull-Text 409-416
  Yin Tangwen; Shan Fu
A Polymorphic Cumulative Learning (PCL) Conception was proposed in order to make it feasible for Digital Pilots (DP) in trying to have the capability to incorporate various forms of learning mechanisms as abundant as human do. An integrated cognitive architecture for analysis of pilot-aircraft dynamic environment (ICA-APADE) which will facilitate aircraft design and evaluation is devised as a framework to make the analysis and implementation of the PCL conception more concrete and practical.
A Context-Aware Adaptation System for Spatial Augmented Reality BIBAKFull-Text 417-425
  Anne Wegerich; Matthias Rötting
From an HCI point of view Augmented Reality (AR) displays have a very specific characteristic. The shown information consists of a virtual and a nearly uncontrollable and cluttering real part. Thus, AR visualizations can be ambiguous, imprecise, or difficult to understand. To avoid this, the presented project shows a systematic method for adapting AR visualizations to the user's needs and perceptual properties. Therefore, we firstly reused and extended established 2D visualization models with AR specific visualization parameters and secondly incorporated known AR specific perception facts to set up an AR smart home control system in a kitchen. The control system involves spatial AR displays (sAR) and makes use of context information like current user tasks or the user's position. The goal is to provide a generic approach of displaying unambiguous AR information at the right time and location without overwhelming the user.
Keywords: Spatial Augmented Reality; Information Visualization; Perception; Evaluation; Context-Awareness; Smart Kitchen Displays
Using Physiological Parameters to Evaluate Operator's Workload in Manual Controlled Rendezvous and Docking (RVD) BIBAKFull-Text 426-435
  Bin Wu; Fang Hou; Zhi Yao; Jianwei Niu; Weifen Huang
How to train the astronauts for complicated operation in spaceflight such as manually controlled rendezvous and docking (RVD) is a common and important question. In this paper, two training methods named Routine method and Confusing method were proposed from astronaut training experience. Two-factor experiment was conducted. The factors are training method and operation difficulty level. Sixteen male subjects participated in this experiment. Eight participants were trained with Routine method and the other eight trained with Confusing method. Physiological parameters, such as respiratory rate, body temperature, heart rate (HR) and heart rate variability (HRV), were adopted as the dependent variables to evaluate operator's workload. Results show that there are significant differences on body temperature, respiratory rate, cubic root of high frequency component (HF) of HRV, normalized low frequency component (LFNu) of HRV, and mean heart rate (HR) (p values are 0.010, 0.000, 0.042, 0.009 and 0.000, respectively) between training methods, while other physiological parameters show no significant difference. Furthermore, only significant difference on body temperature (p value is 0.010) was found between different operation difficulty levels. Therefore, suitable training methods deserve thorough investigation since they have significant effects on the physiological workload of the operators. In conclusion, this work provides insight into the effect of some influencing factors to workload during the RVD procedure, and will benefit the astronaut training in future.
Keywords: Rendezvous and Docking (RVD); Training method; Difficulty level; Physiological parameter; Work load
Task Complexity Related Training Effects on Operation Error of Spaceflight Emergency Task BIBAKFull-Text 436-445
  Yijing Zhang; Bin Wu; Xiang Zhang; Wang Quanpeng; Min Liu
This paper aims to investigate the training effects on the operation errors of Chinese operators when they conduct spaceflight emergent tasks with different complexity. Twenty-eight operators participated in one training experiment and took nine tasks which were divided into three complexity levels based on complexity measures. Six operation errors were recognized based on one engineering classification method. The results showed that training can reduce the errors significantly in practice trails. Different effects were also found on operation errors of tasks with varying complexity. Training did not exert significant influence on some error types, such as time estimation error, which might rely more on the inherent cognitive ability. In test trails, training differentially reduced errors on the high-complexity tasks, but showed no effect on the other tasks. The results implied that future training programs should be designed based on the complexity of tasks and the cognitive characteristics of astronauts.
Keywords: Training effect; Operation error; Task complexity; Emergency tasks; Error classification
The Research of Crew Workload Evaluation Based on Digital Human Model BIBAKFull-Text 446-450
  Yiyuan Zheng; Shan Fu
The present paper presented the requirements of a digital human model used in the research of human factors in flight deck design and the methods to assess the workload, and established one model to evaluate the crew workload based on certain task.
Keywords: Digital Human Model; Human factors; Fuzzy Model Identification

Driver Modeling

A Simulation Environment for Analysis and Optimization of Driver Models BIBAKFull-Text 453-462
  Ola Benderius; Gustav Markkula; Krister Wolff; Mattias Wahde
A simulation environment for evaluation and optimization of driver models is introduced and described. The simulation environment features models of vehicles and drivers, as well as a representation of the traffic environment (roads, buildings etc.). In addition, an optimization framework based on stochastic optimization algorithms has been implemented as an integral part of the simulation environment. Given observed (time series) data of driver behavior and, possibly, vehicle dynamics, the optimization framework can be used for inferring driver model parameters. The simulation environment has been evaluated in two scenarios, one involving emergency braking and one involving a double lane change.
Keywords: driver models; optimization; vehicle simulation; emergency braking; double lane change
Learning the Relevant Percepts of Modular Hierarchical Bayesian Driver Models Using a Bayesian Information Criterion BIBAKFull-Text 463-472
  Mark Eilers; Claus Möbus
Modeling drivers' behavior is essential for the rapid prototyping of error-compensating assistance systems. Various authors proposed control-theoretic and production-system models. Based on psychological studies various perceptual measures (angles, distances, time-to-x-measures) have been proposed for such models. These proposals are partly contradictory and depend on special experimental settings. A general computational vision theory of driving behavior is still pending. We propose the selection of drivers' percepts according to their statistical relevance. In this paper we present a new machine-learning method based on a variant of the Bayesian Information Criterion (BIC) using a parent-child-monitor to obtain minimal sets of percepts which are relevant for drivers' actions in arbitrary scenarios or maneuvers.
Keywords: Probabilistic Driver model; Bayesian Autonomous Driver model; Mixture-of-Behavior model; Bayesian Real-Time-Control; Machine-Learning; Bayesian Information Criterion; Hierarchical Bayesian Models
Impact and Modeling of Driver Behavior Due to Cooperative Assistance Systems BIBAKFull-Text 473-482
  Florian Laquai; Markus Duschl; Gerhard Rigoll
Current developments in Car2X communication technology provide the basis for novel driver assistance systems. To assess the impact of such a system on a group of cars approaching a non moving traffic jam, the driver behavior resulting from a system which supports anticipatory driving is modeled and used in a sub-microscopic traffic simulation. By equipping various percentages of a simulated group of cars with the model, the effect of the system on maximum deceleration and fuel consumption can be assessed. Finally the problems resulting from switching between different driver models are discussed.
Keywords: Driver modeling; Deceleration; Traffic simulation; Anticipatory driving
Predicting the Focus of Attention and Deficits in Situation Awareness with a Modular Hierarchical Bayesian Driver Model BIBAKFull-Text 483-492
  Claus Möbus; Mark Eilers; Hilke Garbe
Situation Awareness (SA) is defined as the perception of elements in the environment within a volume of time and space, the comprehension of their meaning, and the projection of their status in the near future [1]. Lacking SA or having inadequate SA has been identified as one of the primary factors in accidents attributed to human error [2]. In this paper we present a probabilistic machine-learning-based approach for the real-time prediction of the focus of attention and deficits of SA using a Bayesian driver model as a driving monitor. This Bayesian driving monitor generates expectations conditional on the actions of the driver which are treated as evidence in the Bayesian driver model.
Keywords: Focus of attention; deficits in situation awareness; Bayesian autonomous driver model; Bayesian driving monitor; modular hierarchical Bayesian driver model; learning of action-relevant percepts
The Two-Point Visual Control Model of Steering -- New Empirical Evidence BIBAKFull-Text 493-502
  Hendrik Neumann; Barbara Deml
Formal models of human steering behavior can enhance our understanding of perceptual and cognitive processes involved in lateral control. One such model is the two-point visual control model of steering proposed by Salvucci and Gray [8]. An experiment was conducted to test one of its central assumptions, namely that people use information coming from only two locations, a near region about 8m in front of the car and a far region 0,9s down the road, for lane keeping. 42 subjects completed a simulated driving task; road visibility was either unrestricted or reduced according to the assumptions of the two-point model. Additionally, the subjects could either freely choose where to look or had to fixate a target located in the far region. Analysis of steering precision data showed that reduced visibility did not reduce steering precision, thus lending support to the near/far region assumption of the two-point model.
Keywords: Driver Modeling; Lane Keeping; Near/Far Point; Two-Point Visual Control Model of Steering
Automation Effects on Driver's Behaviour When Integrating a PADAS and a Distraction Classifier BIBAFull-Text 503-512
  Fabio Tango; Luca Minin; Raghav Aras; Olivier Pietquin
The FP7 EU project ISi-PADAS aims at conceiving an intelligent system, called PADAS, to support drivers, which intervenes continuously from warning up to automatic braking in the whole longitudinal control of the vehicle. However, such supporting systems can have some unwanted side-effect: due to the presence of automation in the driving task, less attention and reaction are needed by the drivers to intervene in the longitudinal control of the vehicle. Such a paper aims at investigating the effects of the level of automation on drivers, in particular on their Situation Awareness, when the user is supported by a specific PADAS application, integrated with a driver's distraction classifier.
What Is Human? How the Analysis of Brain Dynamics Can Help to Improve and Validate Driver Models BIBAKFull-Text 513-522
  Sebastian Welke; Janna Protzak; Matthias Rötting; Thomas Jürgensohn
To model realistic driver behavior, research interest is growing to develop new approaches that deal with the integration of cognitive processing steps in order to tune the modeled behavior to be more human-like. But what means human-like in this context? This paper deals with the question: How brain dynamics can help to get insights into the internal processes preceding intended driver behavior? Features extracted from electroencephalography (EEG) data can serve as an indicator of such neural activities. This approach will be illustrated by two experiments, dealing with simple button presses and more complex realistic steering maneuvers in an obstacle avoidance task. We found the lateralized readiness potential (LRP) in both experiments preceding button presses (~200ms) and steering maneuvers (~190ms) as a valid neuronal characteristic for motor preparation processes. This data indicates, that a time range about 200ms should be considered between decide and action components to describe driver models more human-like.
Keywords: EEG; LRP; realistic driver modeling
Note: Best Paper Award
Less Driving While Driving? An Approach for the Estimation of Effects of Future Vehicle Automation Systems on Driver Behavior BIBAKFull-Text 523-532
  Bertram Wortelen; Andreas Lüdtke
We present a concept for the automatic estimation of the effect of newly introduced in-vehicle systems on driver's gaze behavior by means of executable cognitive driver models. Structure and properties of the complete system are shown. Particular attention is paid to the underlying model of visual attention which is integrated in the cognitive architecture CASCaS. Different alternatives for the operationalization of human expectancy as one influencing factor of visual attention are presented and discussed. Principal applicability of the model is demonstrated on a laboratory four-instrument visual scanning task. First results are briefly discussed. Future steps are outlined.
Keywords: Cognitive Driver Model; CASCaS; Visual Scanning; In-Vehicle System Evaluation