Effects of Verbal Communication with a Driving Automation System on Driver Situation Awareness
At level 2 of SAE’s driving automation, it is difficult for drivers to continue to monitor a driving automation system and the environment. An important issue to be addressed involves maintaining driver situation awareness to keep the driver in the control loop. In the study, we investigate verbal communication between the driver and system. We hypothesize that the driver can cognitively participate in vehicle operation even if he/she is not physically in the control loop. We use a driving simulator to examine how verbal communication affects driver situation awareness. We compare the following two conditions: (1) talking with the system and (2) not talking with the system during automated driving. Under the condition of talking with the system, the system asks the driver about the peripheral situation and/or vehicle control. The driver is required to respond to the system. In the experiment, two events occur during which the driver is expected to intervene during cruising. We measure the event response time, number of collisions, how the driver maneuvers the vehicle, and subjective usability by administering a questionnaire. The results indicate that the number of collisions are significantly higher under the condition of conversation than under the condition of no conversation. The event response time is significantly longer under the condition of conversation than the condition of no conversation. The aformentioned results indicate that the verbal communication does not improve driver situation awareness. There is no difference in the questionnaire score, and thus the verbal communication does not improve the usability of the driving automation system. The results indicate that drivers can potentially overestimate the extent to which they obtain information about driving situation only through conversation. The results provide important insights for designing systems to support driver situation awareness.
KeywordsDriving automation Situation awareness Verbal communication
The study was supported by JSPS KAKENHI 15H05716.
- 1.Cabinet Office of Japan: SIP Research and development plan of automated driving system. p. 1 (2015)Google Scholar
- 3.Homma R, Wakasugi T, Kodaka K (2016) Basic study on transition to manual from highly automated driving (first report)-effects of drowsiness and non-driving-related-tasks and driver behavior when take-over-requests were provided. Trans Soc Automot Eng Japan 47(2):537–542 (in Japanese)Google Scholar
- 4.SAE (2014): Taxonomy and definitions for terms related to on-ground motor vehicles automated driving systems. J3016Google Scholar
- 5.Inagaki T (2012) Design of human-machine symbiotic – search for “human centered automation”. Morikita Publishing, Tokyo (in Japanese)Google Scholar
- 8.Politis I, Brewster S (2015) Language-based multimodal displays for the handover of control in autonomous car. In: Proceedings of the automotive user interfaces and interactive vehicular applications, pp 3–10Google Scholar
- 9.ISO (2012): Road vehicles — Ergonomic aspects of transport information and control systems — Calibration Tasks for Methods which Assess Demand Due to the Use of In-Vehicle Systems. ISO/TS 14198Google Scholar