Browsing by Author "Mahadevan, Karthik"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Open Access Communicating Awareness and Intent in Autonomous Vehicle-Pedestrian Interaction(2017-10-26) Mahadevan, Karthik; Somanath, Sowmya; Sharlin, EhudDrivers use nonverbal cues such as vehicle speed, eye gaze, and hand gestures to communicate awareness and intent to pedestrians. Conversely, in autonomous vehicles, drivers can be distracted or absent, leaving pedestrians to infer awareness and intent from the vehicle alone. In this paper, we investigate the usefulness of interfaces (beyond vehicle movement) that explicitly communicate awareness and intent of autonomous vehicles to pedestrians, focusing on crosswalk scenarios. We conducted a preliminary study to gain insight on designing interfaces that communicate autonomous vehicle awareness and intent to pedestrians. Based on study outcomes, we developed four prototype interfaces and deployed them in studies involving a Segway and a car. We found interfaces communicating vehicle awareness and intent: (1) can help pedestrians attempting to cross; (2) are not limited to the vehicle and can exist in the environment; and (3) should use a combination of modalities like visual, auditory, and physical.Item Open Access Exploring the Design of Autonomous Vehicle-Pedestrian Interaction(2019-09-12) Mahadevan, Karthik; Sharlin, Ehud; Somanath, SowmyaAutonomous vehicle research today places an emphasis on developing better sensors and algorithms to enable the vehicle to localize itself in the environment, plan routes, and control its movement. Surveying the general public reveals optimism about the technology but also some skepticism about its ability to communicate with vulnerable road users such as pedestrians and cyclists. In today's interaction with vehicles at crosswalks, pedestrians rely on cues originating from the vehicle and the driver. Vehicle cues relate to its kinematics such as speed and stopping distance while driver cues are concerned with communication such as eye gaze and contact, head and body movement, and hand gestures. In autonomous vehicles, however, a driver is not expected to be on-board to provide cues to pedestrians. We attempted to tackle the problem of designing novel ways to facilitate autonomous vehicle-pedestrian interaction at crosswalks. We propose interfaces which communicate an autonomous vehicle's awareness and intent as a means of helping pedestrians make safe crossing decisions. Through our exploration, we make several contributions. First, we propose a design space for building interfaces using different cue modalities and cue locations. From an early exploration of this design space, we prototype interfaces designed to facilitate autonomous vehicle-pedestrian interaction. The interaction between vehicles and pedestrians will become more challenging during the transition period until all vehicles on the road are fully autonomous. During this period which we term mixed traffic, vehicles of varying levels of autonomy will occupy roads, some of which will have drivers, others such as semi-autonomous which may have distracted drivers, and fully autonomous vehicles which may or may not have drivers. To study this problem, we contribute a virtual reality-based pedestrian simulator. Our final contribution relates to the evaluation of interfaces in the real and virtual world where we found their inclusion helped pedestrians make safe crossing decisions.