Designing Interaction and Interfaces for Automated Vehicles: User-Centred Ecological Design and Test


I. ModellingUCEID-The Best of Both Worlds: Combining Ecological Interface Design with User-Centred Design in a Novel Human Factors Method Applied to Automated DrivingINTRODUCTIONWhy Use UCEID?THE UCEID METHODLiterature ReviewData CollectionThematic AnalysisCognitive Work AnalysisConsolidation and Ideas GenerationFiltering and CheckingMETHODOLOGICAL CONSIDERATIONSAdvantagesDisadvantagesTraining and Application TimeToolsSUMMARYACKNOWLEDGEMENTSREFERENCESUsing UCEID to Include the Excluded: An Autonomous Vehicle HMI Inclusive Design Case StudyINTRODUCTIONThis Case Study: Designing an HMI for Level 3/4 Autonomous Car TakeoverAgeing PopulationAgeing and Capability ImpairmentAgeing and Digital Technological Interface CapabilityInclusive DesignAPPROACH AND ACTIVITIESOverview of Explore and Evaluate StageEvaluate Activity: Generation and Processing of Requirements - MethodEvaluate Activity: Generation and Processing of Needs Lists - ResultsCreate Activity: Design Workshop 1InputActivityResultsCreate Activity: Iterative Design DevelopmentEvaluate Activity: Testing with Experts and Users - OverviewCreate Activity: Design Workshop 2InputOutputsCreate Activity: Final Concepts and RefinementDISCUSSION AND CONCLUSIONSACKNOWLEDGEMENTSREFERENCESAPPENDIX: JAGUAR l-PACE HhDAVE HMI AUTONOMOUS STATES AND TRANSITIONS SHOWING INTERFACE CHARACTERISTICSDesigning Autonomy in Cars: A Survey and Two Focus Groups on Driving Flabits of an Inclusive User Group, and Group Attitudes towards Autonomous CarsINTRODUCTIONRELATED WORKUser ViewsInclusivenessSURVEYDescriptionResultsFOCUS GROUPSDescriptionResultsDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCESII. Lo-Fi and Hi-Fi SimulatorsAn Evaluation of Inclusive Dialogue- Based Interfaces for the Takeover of Control in Autonomous CarsINTRODUCTIONDialogue-Based Interfaces DesignedEXPERIMENTParticipantsEquipmentProcedureResultsDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCESThe Design of Takeover Requests in Autonomous Vehicles: Low-Fidelity StudiesINTRODUCTIONInclusive DesignBackground and MotivationThe UCEID: Project Design ContextTheoretical BackgroundDefinition of the Scenario, Aims, and Boundaries of AnalysisInitial Data Collection: Experts' Semi-structured InterviewTechnology Analysis and BenchmarkingThematic Analysis 1Focus GroupsThematic Analysis 2Preferences and User Themes InterpretedWork Domain Analysis (WDA) Abstraction HierarchyControl Task AnalysisSocial Organisation and Cooperation AnalysisDesign WorkshopConcept Refinement and FilteringTHE DESIGN AND FORMATIVE DEVELOPMENT PROCESSAutomotive Takeover Requests (TORs)TOR TimingTOR InterfacesThe Design ConceptsTHE SUMMATIVE TRIALSExperiment 1TrialsResultsDiscussion: Experiment 1ConclusionExperiment 2TrialsResultsDiscussion: Experiment 2CONCLUSIONSACKNOWLEDGEMENTSREFERENCESHow Was It for You? Comparing How Different Levels of Multimodal Situation Awareness Feedback Are Experienced by Human Agents during Transfer of Control of the Driving Task in a Semi- Autonomous VehicleINTRODUCTIONMETHODParticipants and Study DesignEquipmentProcedureMethod of AnalysisRESULTS AND DISCUSSIONWorkloadUsabilityCONCLUSIONACKNOWLEDGEMENTSREFERENCESHuman Driver Post-Takeover Driving Performance in Highly Automated VehiclesINTRODUCTIONMETHODParticipantsExperimental DesignEquipmentProcedureAnalysisRESULTSSpeedSteeringLane DeviationDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCESValidating Operator Event Sequence Diagrams: The Case of Automated Vehicle-to-Human Driver TakeoversINTRODUCTIONOESD DevelopmentSTUDY 1 - VALIDATION OF OESD-MODELLED DRIVER BEHAVIOUR IN A LOWER-FIDELITY DRIVING SIMULATORMethodParticipantsExperimental DesignEquipmentProcedureAnalysisInter-Rater Reliability MethodResultsSTUDY 2 - VALIDATION OF OESD-MODELLED DRIVER BEHAVIOUR IN A HIGHER-FIDELITY DRIVING SIMULATORMethodParticipantsExperimental DesignEquipmentProcedureAnalysisResultsDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCESIII. BenchmarkingBreaking the Cycle of Frustration: Applying Neisser's Perceptual Cycle Model to Drivers of Semi- Autonomous VehiclesINTRODUCTIONThe Perceptual Cycle ModelMETHODParticipantsEquipmentProcedureData AnalysisRESULTS AND DISCUSSION: THREE CASE STUDIES OF DRIVER FRUSTRATIONCase Study 1: 'That was scary ....' - The Risk of an Inappropriate SchemaEvidence of Counter Cycle in Case Study 1Case Study 2: 'Он, I've just done the Distronic again - Impeding Intended ActionsEvidence of Counter Cycle in Case Study 2Case Study 3: 'I think it's green now, ... no it's not!' - Ineffective World InformationEvidence of Counter Cycle in Case Study 3Implications for Interaction DesignEvaluation of Applying PCM to On-Road Concurrent VP DialogueCONCLUSIONSACKNOWLEDGEMENTSREFERENCESSemi-Automated Driving Has Higher Workload and Is Less Acceptable to Drivers than Manual Vehicles: An On-Road Comparison of Three Contemporary SAE Level 2 VehiclesINTRODUCTIONResearch Gap and AimMETHODExperiment DesignParticipantsProcedureData AnalysisRESULTS AND DISCUSSIONSComparisons between Manual and Automated DrivingThe Effects of Complexity in the Driving ConditionThe Effects of Drivers' Prior ExperienceQualitative Investigation of Instances Which May Have Influenced Drivers' Workload and Acceptance in Automated DrivingConsiderations for Designing Driver-Autonomous Vehicle Interaction in Highway EnvironmentConsiderations for Designing Driver-Autonomous Vehicle Interaction in Urban EnvironmentRecommendations for Designing Driver- Autonomous Vehicle InteractionOverall SummaryCONCLUSIONSACKNOWLEDGEMENTSREFERENCESThe Iconography of Vehicle Automation - A Focus Group StudyINTRODUCTIONMETHODParticipantsDesignEquipmentProcedureMethod of AnalysisRESULTSExercise OneIcons Indicating Automation Mode ActiveIcons Indicating Manual Mode or Automation Ending/lnactiveColourSize and Text LabelsExercise TwoExercise ThreeADAS ExperienceDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCESIV. НМI SimulatorCustomisation of Takeover Guidance in Semi-Autonomous VehiclesINTRODUCTIONMETHODParticipantsExperimental DesignEquipmentHMI Design and CustomisationProcedureAnalysisRESULTSSpeedThrottleLane PositionSteering AngleTakeover TimeDISCUSSIONSpeed and ThrottleLane Position and Steering AngleTakeover TimeLimitationsCONCLUSIONSACKNOWLEDGEMENTSREFERENCESEffects of Interface Customisation on Drivers' Takeover Experience in Highly Automated DrivingINTRODUCTIONDriver Experience during TakeoverRelated WorkMETHODParticipantsExperimental DesignEquipmentHMI Design and CustomisationProcedureAnalysisWorkloadUsabilityAcceptanceTrustData AnalysisRESULTS AND DISCUSSIONSWorkloadUsabilityAcceptanceTrustCONCLUSIONACKNOWLEDGEMENTSREFERENCESAccommodating Drivers' Preferences Using a Customised Takeover InterfaceINTRODUCTIONUser-Tailorable InterfacesPurposeMETHODEquipment and Driving SimulatorStudy Interface DesignSelectable Customisation SettingsExperimental DesignProcedurePre-TrialTrialPost-TrialHypothesesHypothesis 1Hypothesis 2Hypothesis 3Hypothesis 4Data AnalysisBinary SettingsOrdinal Settings and Takeover TimeCluster AnalysisPost-Task QuestionnaireParticipantsRESULTSCustomisation SettingsBinaryOrdinalCluster AnalysesTakeover TimePost-Task QuestionnaireDISCUSSIONHypothesesDriver ExperienceLimitations of the StudyCONCLUSION AND FUTURE WORKACKNOWLEDGEMENTSREFERENCESModelling Automation- Human Driver Interactions in Vehicle Takeovers Using OESDsINTRODUCTIONDevelopment of the OESD for Automation- Human Driver TakeoverValidation of MethodsMETHODSParticipantsStudy DesignEquipmentProcedureData Reduction and AnalysisRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTREFERENCESFeedback in Highly Automated Vehicles: What Do Drivers Rely on in Simulated and Real- World Environments?INTRODUCTIONChallenges of Customisable InterfacesWhat is Reliance?Measuring RelianceDevelopment of a New Reliance ScaleEXPERIMENT 1 - SIMULATOR STUDYMethodParticipantsDesignApparatusProcedureMethod of AnalysisResultsEXPERIMENT 2 - ON-ROAD STUDYMethodParticipantsDesignApparatusProcedureMethod of AnalysisResultsDISCUSSIONCONCLUSIONACKNOWLEDGEMENTSREFERENCESV. On-Road and Design GuidelinesCan Allowing Interface Customisation Increase Driver Confidence and Safety Levels in Automated Vehicle TORs?INTRODUCTIONMETHODParticipantsExperimental DesignEquipmentProcedureANALYSISRESULTSThrottleSpeedLongitudinal AccelerationSteering AngleSteering SpeedLateral AccelerationTakeover Protocol TimeTakeover Reaction TimeDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCESEffects of Customisable HMI on Subjective Evaluation of Takeover Experience on the RoadINTRODUCTIONMETHODParticipantsExperimental DesignEquipmentProcedureSample and Data ScreeningData AnalysisRESULTS AND DISCUSSIONSComparison between TrialsWorkloadUsabilityAcceptanceTrustComparison between GendersWorkloadUsabilityAcceptanceTrustComparisons between Ace GroupsWorkloadUsabilityAcceptanceTrustBenefits and Effects of CustomisationInformation Settings for Safe and Timely TakeoverCONCLUSIONACKNOWLEDGEMENTSREFERENCESAccommodating Drivers' Preferences Using a Customised Takeover Interface on UK MotorwaysINTRODUCTIONMETHODSSystem DescriptionStudy VehicleHuman-Machine InterfaceCustomisation SettingsStudy DesignProcedurePre-trialTrialPost-trialParticipantsData AnalysisCustomisation SettingsCluster AnalysisHYPOTHESESRESULTSOverview Customisation SettingsBinary Customisation SettingsOrdinal Customisation SettingsCluster Analysis of Customisation SettingsClustering ParticipantsClustering Binary InterfacesComparing Simulator and On-Road StudyDISCUSSIONHypothesesStudy LimitationsCONCLUSIONACKNOWLEDGEMENTSREFERENCESValidating OESDs in an On-Road Study of Semi-Automated Vehicle-to-Human Driver TakeoversINTRODUCTIONCONSTRUCTION OF OESDsMETHODParticipantsExperimental DesignEquipmentProcedureData reduction and analysisRESULTSDISCUSSIONCONCLUSIONSACKNOWLEDGEMENTSREFERENCESDesign Constraints and Guidelines for the Automation- Human InterfaceDESIGN CONSTRAINTSAllow Driver to Take Control at Any Point during Takeover, Be Sure Hands on Wheel and Feet on PedalsPersonalise Takeover Based on Driver Preferences (And Situation)Allow Option to Complete Non-driving Task (Even If It Means Missed Takeover for Junction/Exit)Allow Sufficient Time for Takeover (Big Individual Differences in Our Studies)Customise Takeover Based on Duration of Being Outside of the Control Loop and Frequency of Takeover (And Context: Road, Weather, Other Road Users, Infrastructure, Signage) - Multimodal Human-Machine Interface (HMI)Querying Situation Awareness of Driver by 'Vehicle Avatar'Make Explicit Who Is in Control of Vehicle - Mode Awareness HMI (Light-up Steering Wheel)Recommended Settings Based on Customer Profiles for CustomisationPre-set Defaults for TakeoverGraduated Alert to Takeover Visual, Audio, Haptic (Escalating)Cue Driver to 'Grab' Steering WheelMake 'Takeover Button' Easy to Access (e.g. Put on Gear Stick)'Repeat' Button and 'OK' Button?Encourage (Facilitate) Visual Checks in Environment and Controls of VehicleDisplay the Vehicle Status and IntentionDriver's HMI Actions Need to Be Clearly Fed BackEyes OutUse System to Aid Manual DrivingSome Level of Personalisation and Setting of LevelsLonger Automated Vehicle-to-Human Driver Takeover in Urban Environment (Compared to Motorway)Takeover Strategy That Guides Visual SearchFeedback to Every Driver Action (Process Needs Adapting to Driver and Situation)ChecklistOption to Request Specific Information of Importance to Driver (If Not in Protocol)Education of Drivers in Rationale and TechniqueTraining (Video) before Being Able to Use Autopilot on RoadsOlder Drivers Do Not Like to Constantly Monitor Automation for Takeover (Timer Only) Trend OnlyDifferences between User Preference and Rankings of UsefulnessCharacteristics of ModalitySynchronise Multimodal Cues - Combining or Single ModalityLongitudinal StudiesDESIGN GUIDELINESDesign Methodological Guidelines (DMG)Interface and Interaction Design Guidelines (IDG)User Trials Guidelines (UTG)ACKNOWLEDGEMENTS
 
Next >