Context

This project explores the unique challenges that drivers face during long road trips oversea and how a voice assistant can help alleviate these challenges to create a more enjoyable and stress-free experience.

Design Goal

Design a tool to help overseas self-driving travelers quickly get familiar with the local situation and arrange trips when they arrive, to reduce anxiety about unfamiliar environments and get the best travel experience.

Introduction

Milla is an intelligent voice assistant design for cars. Through planning intelligent itineraries and emotional driving assistance, Milla (the voice assistant) helps tourists driving overseas to become familiar with local conditions and calmly manage their itinerary to reduce their anxiety towards unfamiliar environments and obtain the best travel experience.

User Research

48.8%
0-39 years old

21.7%
20-29 years old

We searched for relevant research reports and learned that in 2016, the age distribution of overseas self-driving users was mainly 48.8% in the age of 30-39, followed by 20-29 years old, accounting for 21.7%[1].

Therefore, we conducted user interviews with 8 users in this age group who have had experience in self-driving tour in the United States. We asked them about their overseas self-driving experience, what difficulties they encountered, and how to solve them.

[1] White Paper on China's Internet Rental Car Rental Industry 2017[EB/OL]. Analysys International, 2017.

Research Insight

Before the trip

Arrival

During the trip

Due to the scattered information on the ticketing of the ticket hotel, the repeated operation time was unfamiliar with the overseas environment and it was difficult to specify the appropriate itinerary.

Users are uncomfortable with overseas traffic rules and driving styles, which is prone to tension and anxiety.
When encountering the scenery along the way, it is often a glimpse of the flowers, sometimes passing through the places of interest, but do not understand.
After a hard day's journey, it is often tedious to confirm that the next day's itinerary.

Persona

User Journey Map

Design

Product Features

According to our customer research, we believe that they need an intelligent assistant to provide suitable suggestions to help them better plan for trips. Additionally, this assistant should use emotional voice to comfort users and ease their anxieties when they driving.

The assistant has two phases:

Reservation and Departure Phase - App on mobile phones

Gather information:
Airfare, accommodation, booking.
Recommend itinerary:
Recommend attractions and scheduling arrangements based on the travel information and user preference.

Traveling Phase - Voice Assistant on car

Car setting:
Adjust the user’s preferred settings in every rental car intelligently.
Driving tips:
Provide local driving and road information, give warnings when appropriate and explain different traffic signs.
Introduce attractions:
Introduce interesting sites vividly and concisely that passed by during driving.
Schedule itinerary:
Recommend nearby attractions and dining information, adjust scheduling flexibly, and check for the next day's itinerary.

The Conversational Frame

Using a whiteboard, we quickly wrote down the possible dialog scripts that the user and the voice assistant may have during the traveling phase and then sorted out the conversation functional groups accordingly.

The Conversational Script

Opening:

This part should also include some tips for driving:

Car setting:

Driving tips:

Introduce attraction:

Schedule itinerary:

Chatter:

Ending:

User Interfaces

The GUI complements the voice interaction to express the information more completely. For example, maps, traffic signsand other graphic information cannot be completely described by voice and should appear synchronously with voice.

Dash Integration Module

When driving, the users mainly focus on the road, so we believe that only the most important information should appear in the DIM, such as navigation, driving warnings, and traffic signs.

Milla's appearance is always displayed in the upper middle of the DIM, and the new prompt displays below Milla.

Center Panel

The secondary information should appear in the Center Panel, such as car settings, driving tips, and schedule itinerary.

The left side of the panel is used to display the conversation, or to display the state of the car when there is no conversation.

When in the conversation state, the left side transforms into the visual display of the dialogue, and Milla's appearance moves from the upper left corner to the middle, which is more conspicuous.

The icons in the dialogue add the non-verbal layer to the conversation.

When in the Schedule itinerary state, the right side of the panel transforms into a map.

In the visual display of the dialogue, To differ the user's dialogue and Milla's dialogue, I make the thickness and transparency of the font different.

Milla

Context

Design Goal

Introduction

User Research

Research Insight

Persona

User Journey Map

Design

Product Features

The Conversational Frame

The Conversational Script

User Interfaces

User Test

Brain Cognitive Evaluation

Design Patterns

While the user is driving, a voice assistant should provide minimal information to reduce required attention.

A voice assistant should provide user gentle alert before they might make a mistake.

Visual information on the dashboard or HUD can help users quickly digest information while driving.