1 / 22

Comparison of manual vs. speech-based interaction with in-vehicle information systems

Jannette Maciej∗, Mark Vollrath Department of Engineering and Traffic Psychology, Technische Universitat Braunschweig, Gausstrase 23, D-38106 Braunschweig, Germany. Comparison of manual vs. speech-based interaction with in-vehicle information systems. 報告者:楊子群.

roy
Download Presentation

Comparison of manual vs. speech-based interaction with in-vehicle information systems

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Jannette Maciej∗, Mark Vollrath Department of Engineering and Traffic Psychology, Technische Universitat Braunschweig, Gausstrase 23, D-38106 Braunschweig, Germany Comparison of manual vs. speech-based interaction with in-vehicle information systems 報告者:楊子群 Driving Behavior Simulation Lab

  2. Compared operating these either by a touch screen or a speech interface. • A hand-free mobile phone car kit • Music section system • Two different navigation • The two navigation systems differed in the way the speech interface was designed enabling us to also obtain some results on the importance of interface design to driving safety. Objective Driving Behavior Simulation Lab

  3. Method 1.Driving task 2.Secondary tasks Driving Behavior Simulation Lab

  4. Use the Lane Change Task(LCT;Mattes,2003)to evaluate the distraction case by the different in-vehicle information systems(IVIS). • LCT simulation • 1.Straight section of a three-lane road . • 2.Cannot drive Speed faster than 60 km/h. • 3.At some point signs are introduced which become legible at certain distance. • (Indicate driver should change the lane as soon as possible) Method 1.Driving task(1/5) Driving Behavior Simulation Lab

  5. Creating this task was to assess the perdormace in two basic components of the driving task. • Lane-keeping • Lane-changing • The advantage to LCT is to create well-defined level of the driving task difficulty which requires frequent attention of the driver. • Eighteen lane change , that is to say “one lane change every 10s”. Method 1.Driving task(2/5) Driving Behavior Simulation Lab

  6. One trial consists of: random sequence to 18 lane changes (left vs. right,movement across one lane vs.movement across two lane) ---total time:3min. Method Lane-keeping SDLP(Standard deviation of the lateral position) 1.Driving task(3/5) Analysis Lane-changing Reaction time Mean deviation Driving Behavior Simulation Lab

  7. Subjective distraction • Gaze behavior • Drivers rated their perceived distraction into two step. • First step • Five verbal categories ranging form “very little” to “very strong” • Second step • Once more using a 3-point scale ( - / 0 / + ) • This procedure gives reliable data on a 15-point-scale( 5 x 3 ) Method 1.Driving task(4/5) Driving Behavior Simulation Lab

  8. Subjective distraction • Gaze behavior • Following the experiment, using the software INTERACT. • Gaze classified as either: • 1.Looking at the screen of the LCT. • 2.Looking away from the screen towards the secondary task. • Data: • Gaze percentage = away screen(LCT) duration time / total driving time Method 1.Driving task(5/5) Driving Behavior Simulation Lab

  9. Compare manual vs. speech-based four IVIS were examined. 1.Selection same music database. Manual condition >> MP3-Player Speech condition >> Laptop-based prototype music selection system 2.Call by specifying the name and type of the contact. hands-free car kit 3.Navigation system was used for two tasks. (touch screen 、 voice command) (1)Enter specific point-of-interests(POI) in several cities. (2)Enter an address. Method 2.Secondary tasks Driving Behavior Simulation Lab

  10. 30 drivers • 16 male. • 14 female. • Demands of condition • Normal vision (inccorrected). • Age • Rang ed from 19 to 59. • Mean age of 33.2(SD=11.9). • Experiment time • Two to three hours. Subjects Driving Behavior Simulation Lab

  11. Standard PC equipped • Joystick steeringwheel • Gas • Brake pedal Apparatus Driving Behavior Simulation Lab

  12. Within-subjects design • Interface • Manually • Via Speech • Driving the LCT which adds up to eight condition. • Additionally,as a ninthcondition the second navigation system for address entry. • Secondary tasks and order of the interaction mode(manual vs. speech) was randomized. Experimental design Driving Behavior Simulation Lab

  13. Procedure Baseline Music Phone Points-of-interests(POI) Multiple Single Driving Behavior Simulation Lab

  14. The aim of the study is to examine reduces the distraction. • (speech operation compared to manual-visual) • To this aim, a 4×2 (four IVIS×two modes of operation)repeated measures MANOVA. Results Main effects High significant Interaction Driving Behavior Simulation Lab

  15. For the address entry Results For the POIentry Driving Behavior Simulation Lab

  16. For the telephone IVIS Results For the music selection Driving Behavior Simulation Lab

  17. 2 1 Results No advantage of the speech control could be shown Driving Behavior Simulation Lab

  18. Results Significant increase Exception being the point-of-interest entry Driving Behavior Simulation Lab

  19. Significant increase in all tasks Results Due to drivers saw the sign in advance(could not read it) Driving Behavior Simulation Lab

  20. Results 30-40% of the time was used to look at the displays.(mean duration:0.7 to 1s) Substantially reduced by the speech control Driving Behavior Simulation Lab

  21. Subjective distraction was stronger for manual than for speech Results Driving Behavior Simulation Lab

  22. END

More Related