IoT prototype for people and their pets
USeR RESEARCH    Rapid prototype     Arduino     usability testing

Project Summary

Dogs and their owners suffer separation anxiety during the weekday. In the project, we designed an interactive system that allows them to chat and play with each other when the human is not at home.

We used both digital and physical prototypes to help explore UX design in IoT (Internet of things) projects.

My role

During the 3-week project, my major contributions were prototyping and testing.

With lots of practice in previous projects, I was pretty familiar with both mobile-interface prototype and physical prototype. To challenge myself, I volunteered to make the Arduino prototype and focused on how to integrate designs different in different dimensions (IoT and mobile app).

A Story of Petter

Amy has a pet called Petter. Every weekday, Amy has to separate with Petter to go to work. This makes both Amy and Petter upset. They missed each other a lot. Fortunately, they can keep in touch through Holo, an interactive robot that allows them to see and play with each other.

Introducing You, Holo

Holo is an interactive robot that equips multiple screens to present an embodied user's image to dogs. Interactive games are integrated into the robot through which the user can play with their dog remotely to ease the separation anxiety for both humans and dogs.

User Research

Based on our own experience, we know the separation between dogs and their owners can cause anxiety for both. However, we do not know if other dog owners have the same feelings. To support our assumption and concrete the problem scope, user research could be helpful.

Owners inevitably have to leave their dogs at home alone for various reasons like work or holidays. For owners, possible anxiety in making plans and leaving the dog alone at home for a few hours can obstruct their personal lives.

"Separation anxiety is 1 of the most common canine behavior problems and is diagnosed in 20 to 40% of dogs."

"Separation anxiety in pet owners — ranging from a reluctance to leave their dog home for even a few minutes, to a complete inability to travel at all — is a lot more common than we realized."

"Combining physical interaction devices with audio- and video-based system might make such a system more enjoyable for pets."

"Disembodied voices could be problematic to some pets."

James, 41, Empathetic Pet Prioritizer. Pet: Bella.

It’s bad enough worrying about her when I’m at work, but now I’m turning down weekend plans because I feel bad leaving her alone.

Goals & Needs:
Reduce Bella's separation anxiety. Reduce his concern for Bella's wellbeing when leaving her alone. To have more flexibility with his life.

Having to turn down plans due to concern for Bella's possible anxiety.
Occasionally having to leave the office to check on Bella at home.

Katie, 18, Distant, Homesick, Companion. Pet: Millie

I hadn’t prepared myself for what it would be like not seeing her every day, dad sends me pictures of her but it’s just not the same.

Goals & Needs:
Bridging the lost connection with Millie.
The opportunity to spend comforting time with Millie during times of worry.
Decreased homesickness at Purdue.

Having to wait for holidays to come around to spend time with Millie.
Receiving texts and videos of Millie, increasing frustration at the distance.


We are required to use IoT in our design which is an appropriate technology which can benefit the long-distance relationship. With this constraint and research results in our mind, we came up with several ideas as our starting point to design for people and their pets to ease their separation anxiety.

Rapid Prototype

Different from app design, using bodystorming as a design method helps design real/physical world experience. We used diverse foam core, paper, and other available materials to help us think about the experience. Arduino was also used as a part of the IoT project.

Body-storming & product size

Body-storming helps us identify the size and height of the product. We learned that

1. The screen height should be adjustable for different users including children, adults, and elders.

2. The space for games should be designed for different sized dogs to play.

Paper to foam core prototype

We first created a paper prototype to help us confirm the validity of our ideas and the mechanism that we planned to implement. The slide mechanism in our prototype was implemented to simulate the adjustable screens. We used a 9.7 inch iPad as the screen, a foam core and rope to pull it up and down.

Digital Prototype with Arduino

In our system, we wanted the dog to be able to initiate the interaction as well as the owner. The mechanism should to be easy to learn for dogs. Arduino and Adafruit Feather were used to make the mechanism work. In the design scenario, the cushion in the living room works as a trigger. When the dog sits upon the cushion, the sensor will send a text to the owner through the Adafruit Feather WiFi connection to tell him/her that the dog is calling. Then, the owner can turn on the video chat and interact with the dog.

Learning by Doing

After making the prototype, the first thing we did is going through the whole interaction to identify obvious usability problems.

We acted it out rather than just conducting a heuristic evaluation. It turns out that the acting was valuable and helped us find problems as follows.

Slide to answer function on the Ipad

An iPad was used to simulate the screen. However, the dog cannot slide on the iPad and pick up the video chat. This is a valuable insight because none of us thought about this. We were too familiar with an iPad and swipe to answer a video chat is natural. Dogs cannot do this.

Camera Angle

We only considered the height of the camera. However, we cannot assume the dog never moves. Thus, the angle of the screen should also be designed.

Testing Feedbacks

We were not sure about how the system works with real users since it is more complicated than designing a digital experience. Three quick user tests helped us gain more insights.

Finding dog-owners and convince them to bring their dogs to test is ideal but not feasible since we do not have enough time. We decided to test the human side of the design rather than the dog side. Thus, one of us acted like a dog to cooperate with users to complete the test.

We focused more on what users were concerned with, and their thoughts on potential functionality and uses when they are interacting with the prototype.

One user worried that calling the dog will disturb it if it is sleeping. This inspired us that we would look to silence the calling alert sound, and perhaps dim the brightness of the screen.

We tested with the tug of war and also told users of a possible fetch game. One user's concern was that their dog may not come back after the ball was thrown. We could look to attach a line to the ball in the fetch game to counter this concern.

One user worried that playing with the dog 2 or 3 times a day will make the dog more anxious when she is away because it will increase the dependence of the dog. This is one point we are not sure as to the extent of the problem and would require robust research.


The design becomes more challenging when we have to consider both digital and physical elements. It has a huge difference with just making a visually appealed prototype on screen.

As I mentioned, I chose to play around with Arduino to challenge myself. However, it turns out the Arduino does not affect the whole experience too much. It is just a tool to achieve our goal for long-distance interaction. Methods like body storming are actually more useful in such projects. Even it is not perfect, acting like a dog can still help us identify problems in the experience that we could not found in a human's perspective, for example, the angle of the camera, the swipe function on the screen.

With an urban designing background, I found that I like to design in this way. I am good at making prototypes fast and making design decisions based on the rapid prototypes. Learning by doing even gives me the confidence to believe my design will actually work.