The success of Anki Vector in research in Human Robotics Interaction (HRI)

One of the few successes that Anki robots, Vector and Cozmo, (now owned by Digital Dream Labs) have had is in the exploration of Human…

One of the few successes that Anki robots, Vector and Cozmo, (now owned by Digital Dream Labs) have had is in the exploration of Human Robot Interaction (HRI). HRI is a multi-discliplinary research field that explores how humans perceive robots and interact with them. This is a very important research area because the success of robotics in everyday environments such as the home, schools, or shopping malls, depends on how well humans deal with the presence of robots and cooperate with them on tasks that the robots are designed to let the humans accomplish. Having a healthy and friendly interaction between the human and the robot is the key for successful deployment of robots closer to our daily lives.

It turns out that both Vector and Cozmo are receiving significant adoption in the community of HRI researchers. The recently concluded ACM/IEEE Conference on Human Robot Interaction (HRI 2021) has many examples of research using Vector and Cozmo. This article summarizes some interesting pieces of research which were published in HRI 2021.

Social Interactions via a robot and app:

The authors of VectorConnect discuss how they developed the VectorConnect App, the challenges they encountered and the lessons they learnt in the process in this article. VectorConnect allows children living in different households to play with a common robot and interact socially with the help of the VectorConnect app. A more detailed evaluation of VectorConnect is available in this Programming Robots article.

The most interesting part of VectorConnect’s HRI paper is a frank discussion of the issues that the authors had to work with while trying to deploy this app; such as navigating the COVID-19 pandemic, technical issues working with the Vector APIs, dealing with privacy issues, and navigating with university processes in getting the application certified.

Helper’s High:

Another interesting piece of research from Bristol University UK, investigates if helping a robot pet (such as Vector) accomplish its tasks help enhance the mood of humans, also known as Helper’s high. This study involved three participants, two of which reported that they felt better after interacting with Vector. One of the participants even indicated that she would like to have a pet, but didn’t see the need to having one if she could play with Vector all day. However, the third participant didn’t enjoy the company of Vector. One interesting difference in the living situation of the three participants was the two participants who liked interacting with Vector, where living by themselves; while the third participant was living with friends. When the participants felt lonely, the first two participants would interact with Vector for company, while the third participant would talk to his friends.

This indicates that differences in living situations would create differences in the way humans respond to a robot in their presence.

Fair allocations for tasks:

In another interesting piece of work, researchers from Cornell University explore fairness considerations that robots such as Vector must keep in mind while working an interacting with teams of humans. The study designes an environment where a Vector robot helps a team of two humans with a manufacturing problem, where Vector’s job is to assign the pieces to each human. The investigation tries to understand how Vector’s decisions influences the behavior and performance of the two team members.

Impact of verbal interactions:

Apart from these studies, there are many other researchers that have employed Anki Cozmo in their work. An interesting study from Tokyo Institute of Technology looks into whether verbal interactions of a robot creates a better relationship with humans, and makes it more likely for the human to help the robot in its task. 24 paid students were asked to work with a interactive and non-interactive Anki Cozmo for 20 minutes, help the Cozmo in its tasks, and then answer questionaires on how they felt about the robot.

The study shows that verbal interactions with the human lead to a better relationship between the human and the robot. However, there was not any conclusive evidence to suggest that an interactive robot receives better help from humans.

It is clear that Anki Vector and Cozmo have a great potential in research, specifically in HRI. The main reasons seem to be the relatively cheap cost of the robot, a complete set of APIs which lets one program the robots in any desired way, and the ease of deploying these robots and having humans interact with them. We look forward to seeing more interesting pieces of research using these robots in the coming years.

If you have any questions, or thoughts, please leave it in the comments below. Please follow my publication: “Programming Robots” for more interesting articles. I also have an online course to teach AI with the help of Vector available at: I will feel honored to have you as a student.