My role: User experience researcher and designer: co-designed the experiment, designed interview questionnaires, conducted interviews with parents and children, analyzed data, brainstormed personas, scenarios, and design solutions, tested the prototypes, moderated co-design sessions with users, and designed the high-fidelity prototype of the parent app.
Study outcomes: The results of this study were submitted as a paper and accepted to the Late Breaking Works of CHI 2020. The paper has been published as CHI Extended Abstracts in the ACM Digital Library.
"Otto" phone and tablet app home screens
The Challenge
The rise of the autonomous transportation system is reshaping the future of all vehicles, including school buses. While children may soon be riding autonomous buses (AB), there is uncertainty about whether parents and children are ready for this change.
Little research is reported on children’s behavior in the ABs under real road conditions and on improving parents’ trust in leaving their children alone in ABs. Our team saw an opportunity to answer the following question:
How can we design ABs suitable for unaccompanied children so that the parents can trust the systems?
Objectives
• Understand children’s behaviors, desires, and concerns about ABs
• Understand parents’ feelings and concerns about letting their children ride on an AB
• Design a system that accommodates the needs of children and parents.
The focus was on three main concepts – trust, perceived safety, and perceived usefulness.
Research and Design tools
Figma | Adobe CC (Illustrator, Photoshop, Premiere Rush) | Balsamiq | Otter.ai
Design Process
Empathize
The team conducted a study using a Wizard-of-Oz technique to observe how children behaved in an SUV modified to simulate an "autonomous vehicle."
A foam board was used to separate the front and back seats and a GoPro camera was installed in the middle of the board to record the children’s behavior during a 15-minute ride (Figure 1).
Field observations were conducted with 4 elementary school children between the ages of 6 and 10 so that the illusion of an AB could be easily created.
Experiment Procedure
• Participants were requested to spend 10 minutes sitting in a car to familiarize themselves with the surroundings.
• Then they were taken around the neighborhood for a 15-minute ride, followed by a short interview. Participants were instructed to use the "Robot Stop" command if they felt scared at any time during the ride. If activated, the car would stop, and the team would be alerted to provide assistance.
The team conducted 3 rides, with two children in the vehicle each time.
Figure 1: A photo of an SUV that was transformed into an "autonomous vehicle"
Figure 2: Two children seated in the car's backseat
Following the rides, I conducted interviews with the parents to explore their perceptions of ABs.
The focus was on the following objectives:
• how children behave in the study compared to the regular rides with their parents
• how much parents know about autonomous vehicles
• whether they are ready to let their children ride in autonomous vehicles
• the features parents want to see in such systems.
• how children behave in the study compared to the regular rides with their parents
• how much parents know about autonomous vehicles
• whether they are ready to let their children ride in autonomous vehicles
• the features parents want to see in such systems.
Video of the data collection and analysis process
Define
When analyzing data, we discovered a lack of unanimity among parents regarding the idea of their children being transported in an autonomous vehicle. There were also conflicting perspectives regarding the age or maturity levels at which parents would be comfortable allowing the children to ride in a vehicle without adult supervision.
The valuable insights gained from the interviews helped us form our problem statements and hypothesis centered around the goal of enhancing trust in autonomous vehicle technology.
Figure 3: A photo of an affinity diagram
Reviewing video recordings from the GoPro camera provided the team with an opportunity to observe patterns in children's behavior and identify potentially risky situations as they interacted with the autonomous system.
Our notes were regrouped multiple times during the affinity mapping process until the main categories were identified: trust, entertainment, behavior, and desires.
Personas & Scenarios
The notes from the field studies helped me refine our persons and their goals. After some brainstorming, the team reached a consensus to prioritize three personas, each with distinct needs and goals: Poppy, Carl, and Karen & Bill.
Figure 4: Persona 1 - Poppy
persona 01: Poppy is a 7-year-old cheerful and active child who enjoys interacting with other children, being at the center of attention, making new friends, and expressing her opinions on specific subjects. She is naturally curious, often posing numerous questions, and can easily become impatient with others. Due to her active and strong personality, she may experience frustration and boredom during bus rides if not sufficiently engaged and entertained, leading to potential disruptive behavior.
Hobbies
•Make rubber band bracelets.
•Play with dolls, interactive games, and puzzles.
•Make rubber band bracelets.
•Play with dolls, interactive games, and puzzles.
Goals
•Stay active and entertained during the ride.
•Demonstrate to her friends that she has learned something new and share the knowledge with them.
•Stay active and entertained during the ride.
•Demonstrate to her friends that she has learned something new and share the knowledge with them.
Frustrations
•Does not enjoy reading books while on the road.
•Feels frustrated being confined to the vehicle seat.
•Does not enjoy reading books while on the road.
•Feels frustrated being confined to the vehicle seat.
Figure 5: Persona 2 - Carl
persona 02: Carl is a 9-year-old reserved child with a quiet, somewhat shy, and thoughtful personality. While friendly, he tends to observe rather than actively engage with others. He spends considerable time in front of screens, enjoying electronic books, audiobooks, and podcasts and exploring the internet for new knowledge. He is very close to his parents, seeking their advice and engaging in conversations with them on various topics.
Hobbies
•Likes playing video games and Legos.
•His preferred pastimes are eating pizza, watching his favorite TV shows, and listening to audiobooks and podcasts.
•Likes playing video games and Legos.
•His preferred pastimes are eating pizza, watching his favorite TV shows, and listening to audiobooks and podcasts.
Goals
•Stay engaged and connected during the ride.
•Easily get emotional support when needed and feel safe on the bus.
•Stay engaged and connected during the ride.
•Easily get emotional support when needed and feel safe on the bus.
Frustrations
•Long rides without his parents can make him nervous.
•Feels frustrated when technology malfunctions or slows down, unnecessarily complicating tasks.
•Long rides without his parents can make him nervous.
•Feels frustrated when technology malfunctions or slows down, unnecessarily complicating tasks.
Figure 6: Persona 3 - Karen & Bill
persona 03: Karen and Bill, upper-middle-class working parents, have the financial means to support their children's childcare needs while they are away or working long hours. However, their primary concern is always the safety of their children. When traveling for work, they often prefer leaving their children in the care of grandparents or trusted family and friends in town. Ensuring their children safely reach their extracurricular activities or various destinations poses a challenge. The new autonomous bus systems provide them with peace of mind that their children can safely reach their destinations.
Hobbies
•Play tennis together and travel for pleasure as a family.
•Enjoy outdoor activities with their children.
•Play tennis together and travel for pleasure as a family.
•Enjoy outdoor activities with their children.
Goals
•Keep their children safe and happy.
•Stay informed about the whereabouts of their children.
•Keep their children safe and happy.
•Stay informed about the whereabouts of their children.
Frustrations
•Balance work responsibilities with family life.
•Leave their children alone to ride with others.
•Balance work responsibilities with family life.
•Leave their children alone to ride with others.
The scenarios were created to help us identify what might happen during a ride, how the design can handle these scenarios, and what functions and features the system should support.
It was agreed that the key activities the system should support include:
•Ensure a secure boarding process on the bus
•Enable communication with parents
•Offer emotional support
•Remotely monitor children to intervene when necessary
•Provide diverse entertainment options such as listening to a book or engaging in educational games
•Notify parents upon the child's arrival at the destination.
scenario 01: Poppy is at the bus stop with her mother, fidgeting with excitement as she awaits the arrival of the AB. Finally, with a gentle hum, the AB glided to a stop in front of them. Poppy eagerly scans her ID card on the sleek interface, and with a soft chime, the door slides open. She hops onto the bus, finding her usual seat by the window. The ride starts smoothly, and Poppy watches the familiar scenery pass by, lost in her thoughts. However, her reverie is soon interrupted by the arrival of her friend Lily, who, as usual, is full of energy. Lily slides into the seat next to Poppy, bubbling with excitement.
As the bus continues its journey, an innocent conversation between Poppy and Lily takes a sudden turn. A disagreement escalates into a full-fledged argument, with raised voices and passionate gestures. Poppy, frustrated, starts throwing hastily folded paper at Lily, attempting to convey her point without more words.
An agent in the AB's central monitoring system detects the rising tension. The agent, trained to ensure the safety and well-being of the young passengers, initiates a call. The agent tries to talk about the misunderstanding and find a resolution so both can have a pleasant ride. Misunderstandings and differences are addressed, and a compromise is reached. Poppy and Lily, now understanding each other better, relax in their seats and start chatting amicably. The AB approaches the school, its doors whooshing open. Exiting the bus, Poppy and Lily promptly break into a run towards the school.
The morning's conflict was resolved with the help of advanced technology and a caring agent (Figure 7).
Figure 7: My sketching of Scenario 1
scenario 02: Carl's parents, Karen and Bill, are running late for their flight, so they use the autonomous shuttle mobile application to set up a ride for him. He is nervous to take the bus but his mom, Karen, has assured him that everything will be ok. Carl's heart races as he stands at the curb, watching the autonomous shuttle approach. Taking a deep breath, Carl steps onto the bus, feeling a mix of excitement and anxiety.
As the bus begins to move, Carl gazes out of the window, watching the familiar buildings shrink in the distance. He is nervous and he wants to hear his mom’s voice. Remembering his mom's words, he reaches for the device beside him. With a tap on the screen, he initiates a call, "Call Karen." On Karen's phone at the airport, a notification chimes, and she answers the call. The screens display their faces, and Carl's worried expression meets Karen's and Bill's comforting smiles. After a brief conversation, Karen signs off, encouraging Carl to relax and read his favorite book. Now Carl feels at ease. He decides to immerse himself in a familiar comfort, "Read me Captain Underpants." The automated voice of the storytelling app narrates the hilarious adventures of Captain Underpants. Carl's grip on the handle slowly loosens as he enjoys the antics of the beloved character. Meanwhile, Karen occasionally checks the trip progress on her phone. She smiles as the system sends a confirmation message: "Carl has safely arrived." Grandma, eagerly awaiting Carl, has just verified the message.
Carl couldn't help but appreciate the seamless and safe journey the autonomous shuttle had provided. The blend of technology and a mother's comforting voice had made his adventure not just tolerable, but enjoyable.
Ideate
The team combined brainstorming with group scribble sketching on a whiteboard to generate ideas (Figures 8 & 9).
Each team member selected the top five ideas from the board to move to the next round for brainstorming.
During the second round of brainstorming, ideas were selected and combined into full system solutions - a smartphone and tablet application, named Otto system, for parents and children to use in AB rides.
Figure 8: Ideas sketched out for the monitoring process
Figure 9: Scribbling sketching of entertainment options
Findings from the brainstorming session:
•The design should prioritize safety, entertainment, and smooth interaction.
•Using edutainment tablets can help achieve both safety and entertainment.
•The system must enable communication with support staff, parents, the school, and emergency services as needed.
•Both children and parents should have an easy and secure way of communication through video and chat.
•Parents should be able to monitor their child's location and activities.
Prototype & Test
After sketching the ideas and wireframing the child and parent interfaces, the low-fidelity vertical prototypes were developed. The prototypes were brought to testing and co-design sessions with parents and children.
Participants were asked to think out loud while clicking through the interfaces to use different features. Participants were also provided with design printouts and encouraged to express their thoughts on areas for improvement. The feedback gathered from think-aloud and co-design sessions helped to improve the designs.
Figure 10: Prototype testing
Findings from testing sessions:
A. OttoParent should instill trust while providing simple functionality. The system should enable parents to monitor their children inside the vehicle, contact school administration and emergency personnel, communicate with the children, and get updates about the status of pick-up and drop-off, the physical location of the car, route changes, and other relevant updates. The parents need to feel that the system can take care of their children if and when they are unavailable. Push notifications from the app to the parent's phone are essential to support communication between the system and the parent and contribute to building trust.
B. The Otto app must entertain children while providing easily accessible communication options to connect with parents or emergency personnel. If children are not entertained, they are more likely to cause trouble with other children or experience negative emotions.
OttoParent
Parents first encounter a title page with log-in/registration options. They are guided to create an account and input their child's/children's information. Within the child profile, parents can engage in text or video chats with their child, track the bus, or contact support.
OttoParent Wireframes
Figure 11: Parent/guardian interface Version 1 (logging in)
Figure 12: Parent/guardian interface Version 2 (logging in)
The bus tracking screen displays the route map, bus location, and estimated arrival time.
The Contact Support screen provides both emergency and problem-reporting options. The emergency option establishes a direct connection between the parent and authorities, while the "Report a problem" section enables them to submit text feedback for resolution.
Finally, push notifications provide updates, notifying parents of events such as their child's safe arrival or bus delays.
Figure 13: Parent/guardian interface Version 1 (monitoring and chatting)
Figure 14: Parent/guardian interface Version 2 (monitoring and chatting)
OttoParent Mockups & Prototype
Figure 15: OttoParent mobile application screens
Figure 16: OttoParent log-in, ride request, and notification screens
Figure 17: Messaging, video calling, progress monitoring, and feedback screens
The guardian is logging in to review notifications for any recent updates.
Keeping an eye on the child and monitoring AB's progress.
Initiating a video call with the child for a brief chat.
Otto
The Otto app prioritizes a visually engaging interface to entertain children while providing easy communication with caregivers and emergency services. Recognizing that disengagement may lead to disruptive behavior or negative emotions, such as loneliness and fear, the app ensures efficient communication channels for children to reach out to caregivers and emergency responders.
Entertainment for children encompasses various activities like games, books, videos, tracking the trip's progress, and chatting with others. The system aims to cater to these interests while encouraging interaction among passengers. An animated character named Otto acts as a companion, guiding children and facilitating communication.
Figure 18: Child Interface Version 1
Upon entering the vehicle, children encounter the homepage featuring Otto, a visual representation of fellow bus riders, and navigation options. Interaction can occur through direct communication with Otto or by clicking on the screen. Exploring the activities of other bus riders allows a child to join them in those activities. Opting for "Let’s Have Fun" leads to games, books, videos, and homework choices.
Figure 19: Child Interface Version 2
The "My Map" feature allows the child to monitor the bus's location on a map, showing the route, destination, and trip progress. The "I Need Help" option provides ways to contact the school, parents/caregivers, or emergency services, with a two-click step for the emergency option to prevent false alarms.
Figure 20: Child Interface Version 3
Figure 21: Child interface Version 3 entertainment screens
DESIGN SPECIFICATIONS
The system encompasses three primary touchpoints:
a) Otto - An onboard system tailored for children, offering entertainment, instruction, and emergency communication channels with the appropriate personnel. It requires access to video and educational libraries and the ability to receive prompts from OttoParent settings for parental control. Otto application is intended to function using a tablet interface.
b) OttoParent - A phone application designed for parents to monitor the bus and their children's whereabouts.
c) Ingress/Egress - A comprehensive hardware/software solution ensuring accurate pick-up and drop-off of the correct child at the designated location.
Approval from school administrators and trained adult monitors is important to handle video chats with passengers and parents. The system also requires updates before the start of each semester to incorporate changes in the library content (books/games/videos).
CONCLUSION
Based on the findings, parents were primarily focused on ensuring the safety of their children, while the children aimed to avoid boredom and fear. However, there were shared concerns about potential emergencies of varying severity, requiring a solution that could encompass a diverse range of responses.
Our findings ultimately came down to a series of requirements based on our study and existing literature. The objective was to keep children entertained while securely seated. The optimal solution involved the use of stowable edutainment tablets that facilitated contact with Otto support staff, parents/caregivers, the school, and emergency services when necessary. For parents/caregivers, the system enabled location tracking, activity monitoring, and video communication with their children.
Children or passengers showed minimal concern about the safety of the autonomous bus. The efforts to enhance trust and safety perceptions were predominantly directed toward parents. The Otto system effectively tackled issues of trust, safety, and functionality, making autonomous school buses well-suited for children. This approach aimed to foster confidence in automated school buses, ultimately leading to widespread acceptance of this innovative school bus system.
Limitations
The team recognized a challenge in working with participants who had no experience with ABs, as they primarily relied on mental imagery to envision the experience of riding in one. Although children were in a simulated AV (using the Wizard-of-Oz method) and perceived it as a "robotic car," maintaining the subject's illusion of autonomy throughout the experiment was challenging. It was equally difficult to assess a child's genuine comfort level in the car.
Moreover, the study involved only four elementary school children to create the belief that it was an AV. Future research should broaden its scope to include other age groups, testing the Otto system in AVs.
Finally, there is a need for further research to clarify children's understanding of the appropriate level of assistance as there was ambiguity regarding who should be contacted for specific emergencies.
