A conversational interface for autonomous ridesharing vehicles.


competitive analysis -> requirements gathering -> scenario writing -> conversational modeling -> wizard-of-oz testing -> medium-fidelity prototyping -> video production


Jia Liu, Xueting Zhang


2017 / 2 months

Problem Statement

In a future in which autonomous vehicles become ubiquitous, there will be a need to create a seamless ride-to-ride experience for users of ride-sharing services such as Uber an Lyft. In the absence of a friendly human, the need to make the users comfortable falls on the shoulders of the interface mediating the user's experience.


Rover is a multimodal user inferface with voice, text, and visuals. It is designed for an autonomous-vehicle enabled future in which most people prefer using a ride-sharing service to owning a car. Rover is is male, with a full, deep voice. He is a friendly and proactive assistant. He stays with users across rides, and he is able to access maps, traffic informaiton, dining options, events, and transportation schedules.. He remembers past conversations and references recent ones. Rover's overarching goal is to make passengers feel more at ease as they order and ride cabs.

Selected Assets

A selection of assets from the prototyping process are presented here.

[Figure 1] A competitive analysis showing the other conversational interfaces in the market at the project's inception.

[Figure 2] The fundamental requirements for a conversation-first interface, as determined from our requirements-gathering process.

[Figure 3] A simple model showing a happy path through a vonversation with Rover.

[Figure 4] A simple outline of Rover's desired personality.

[Figure 5] A model of an end-to-end conversation transacted via Rover.

[Figure 6] A concept video highlighting the experience of riding a cab with Rover.