[DADeR/Materiality] “Motive Intentions: Communication through Movement” – Vivian de Koning

Concepts: materiality, address, kinaesthesia, intention
Case Studies: Roomba (iRobot), Spot (Boston Dynamics)

Abstract

Movement cannot be imply meaning on its own; rather, it is the surrounding environment that gives meaning to it. Based on the concepts of “address” and “kinaesthesia”, this paper argues that robots move to make sense of their environment, in order to best traverse it, and learn in the traversing what their environment is. Due to the distinctly non-human looks of both the Roomba and Spot, these robots cannot move in a way that is human. They can, however, still move in ways that lead to affection or affinity for these objects. To express certain intentions through a robot, one must design its movements based on their materiality rather than blindly mimicking humans.

In this DADeR paper, I am going to focus on the materiality of the robot and its kinaesthetic possibilities. My main piece of advice would be to focus on the actual material properties of the robot that allows for a convincing communication of the concept of the robot, instead of “forcing” the robot into an idea, as it were. My focus on this paper will be on the hardware of the robots and how their movements can be a valuable site of meaning. In order to illustrate my point I will be discussing two types of robots, first of all, Spot from Boston Dynamics, and second, the Roomba. I have chosen for these two robots, since they are built for very different purposes and thus move in very different ways. One of my main sources of inspiration for this paper was Peter Eckersall’s description of Kris Verdonkk’s work, namely that “Verdonck’s work aims to “ask what […] objects and machines can tell us about what they do.”[1]

In order to discuss how robots’ movements can communicate different meanings, I will first discuss the terms “address” and “kinaesthesia” and how they been used in theatre and performance studies, in order to argue that the robot and its movements forces its spectator to relate to it in a certain way. After this, I will briefly introduce my case studies. Next, I will discuss several texts about moving non-living materials, such as puppets and robots. This way, I will sketch a clearer view of how the movements of puppets and robots are related, which gives me the ability to discuss my case studies. And lastly, I will look back on what the movement of robots tells us about what they can do.

Address and kinaesthesia

Theatre and Performance researcher Maaike Bleeker and interaction design researcher Marco Rozendaal have written “Dramaturgy for Devices” together, in which they explain and explore how concepts from the theatre can function as analytical tools in the design of robots. I specifically want to look at the term “address” here. They write, “modes of address affect how the one being addressed is invited to understand what is shown and done, is invited to sympathize or not and so forth.”[2] The address towards the spectator thus influences how the spectator will interpret and position themself in relation to what it is that is seen. In performance studies, this term has long been used in order to discuss how what happens on a stage, or the performance space, in relation to the spectator. In this paper, I will look at the way the robots’ movement communicate meaning to its spectators, or users.

Susan Leigh Foster explains in her book Choreographing Empathy how the term kinaesthesia has been used in performance studies, as well as its application in medicine and neurobiology. She states that in neurobiology, the term had mostly been replaced by proprioception. In dance theory, however, the term has been linked to emotional experience by John Martin.[3] Furthermore, in The Cambridge Companion to Performance Studies, she writes how Martin argues that the viewer can feel similar sensations to the (dancing) body that is seen, that “viewers’ bodies, even in their seated stillness, nonetheless feel what the dancing body is feeling.”[4] It is not only the physical body that the viewer can feel, but when taking the proprioceptive angle into account, James J. Gibson understands movement as a way to understand the environment you are in.[5] This, of course, can lead directly back to robots, as both humans and robots use sensors in order to orient themselves in space. It also leads directly to my case studies, as both robots move and have to make sense of their environment in order to best traverse it, and learn in the traversing what their environment is.

Spot and the Roomba

Boston Dynamics’ Spot is an interesting case study, because Boston Dynamics’ work is entirely based on motion and how to make the robots move around in the environment best. These robots are interesting on several levels. First of all, technologically they are very impressive, they are able to perform quite acrobatic feats, they have a large range of motion. However, more interesting is how they are used. In one promotional video, they are programmed to perform a dance to Do You Love Me?. The Spot model is visible, as well as some more humanoid robots. Their movements are smooth and lifelike, almost suggestive of consciousness. The humanoid robots are programmed to move in a way that is more reminiscent of humans performing ballet. The Spot model is more springy, bouncing around and moving the arm that is suggestive of a head around. All limbs are thus engaged with the dance, not in a motion that is easily recognised as human, but rather what an animated dog might dance like, this is further suggested by the very canine name, “Spot”. While the Boston Dynamics website suggests that Spot is developed to assist people in menial jobs, such as construction and mining, Spot has already been used in military exercises, as posted on twitter by the Académie militaire de Saint-Cyr Coëtquidan.[6]

Figure 1: Picture from Académie militaire de Saint-Cyr Coëtquidan’s Twitter

In one video titled The Evolution of Boston Dynamics, the company and its products are shown through the years.[7] While there is a focus on the design of the robot, there is less said about its applications, in what I can only assume to be an attempt to keep a friendly reputation. Spot is shown to be increasingly able to handle being pushed and navigating in rough terrain.

The Roomba is a circular robot vacuum cleaner, which senses its way around the house in order to get rid of dust and dirt. It can move horizontally and somewhat vertically, to get on a carpet for example. The Roomba has in some cases almost become a pet, in which the owner of this new pet relationship can compare the Roomba to the actual pets (if there are any). There is something that is almost a little pathetic in an entirely encapturing way about the Roomba. Its bumps into the wall can seem silly and perhaps childlike. Over time, however, the Roomba learns the layout of the room, with only small deviations in chairs that are scooped back. Its movement through the room therefore evolves, from a newborn creature to an adept independent hoover. As a user, seeing this development can seem like a little entity growing up, which can develop into an affectionate relationship. You might not expect it, but iRobot, the company behind the by now commercial robot, originally made this robot for military purposes as well. Even today, if you are on active duty in the US Armed Forces, you can get a military discount.[8]

Puppetry and robotics

In order to better understand robot movement, I am also looking at puppetry, as puppeteers are also involved in the steering of movement in non-living things. A big difference however is of course that puppets are in direct contact with the one steering them and that automotive robots are reliant on their own hardware to move around. One study in which these two different areas come in contact is in The Routledge Companion to Puppetry and Material Performance, as in one chapter, Elizabeth Ann Jochum and Todd Murphey have researched the possibilities of controlling puppets with robots, and how to best use mechanics to emulate the liveliness of the marionette. One of the main points they make is that puppeteers have managed to create a vocabulary of movement for marionettes, not one that convinces its audience that it is human, but rather that it lives. At the same time, they state that puppeteers have to work with the physical dynamics of the puppet and thus have to reach a compromise in order to create expressive characters. [9] As the Roomba is a disk on wheels, there is not much human movement to imitate. And while the goal of the robot might not have been creating a lively character, it is exactly by its movement in learning the environment that this character is established. Similar to Foster, Jochum and Murphey comment on the relation between motion and this character.

“In both puppetry and robotics, expressive movement is an integral aspect of mimesis, influencing how deftly the illusion of life is created and sustained. Joseph Roach observes that “expressive movement is becoming a lingua franca, the basis of a newly experienced affective cognition and corporal empathy.””[10]

Movement is thus a way that a robot can endear itself to humans. Due to the distinctly non-human looks of both the Roomba and Spot, these robots cannot move in a way that is human. They can rather move in ways that lead to affection or affinity for these objects.

However, there are teams that do attempt to recreate human movement or looks, such as Sophia.[11] Referencing Masahiro Mori, Jochum and Murphey explain that this goal of imitation can have an unsettling effect, entering Mori’s “uncanny valley”.[12] And it is exactly this tension that is also present in working with robots: how do you work with the physical properties of a robot, not to simulate human movement, but in order to make the robot a convincing character? That is not to say that this uncanny valley is inherently bad, but it could make the interaction less “pleasant” for the user. It all depends on what you want to achieve with the robot whether exploring this uncanny valley could benefit the work.

In “Designing Robots Creatively”, Mari Velonaki and David Rye describe four art projects with robots, with a focus on the interaction with the robot. In the fish-bird project, they describe why was chosen for using a wheelchair in an art installation; because it is a kinetic object that both reminds one of the absence, as well as the presence of a person.[13] In this way, they are explicitly engaging the robot design to the user/spectator’s experience, and how the robot’s movements, in combination with its design, communicates the message they want to send. Later in the text, Veloraki describes how she had to create some “reference points” for another work in order to better communicate its allowances and intents to people.[14] In this chapter, they argue that it is “important to understand and account for the context of the interaction and the culture within which it will take place.”[15] I interpret this to mean that when using robots, they can never be seen as separate from their environment. However, I do not think that robots will always be bound to this environment. Robots are made with a certain purpose, but whether this intention is “honoured” is another question. The Roomba learns your floor plan in order to clean best, I can think of other reasons why it would want to know a floor plan. Spot is able to move through rough terrain; this is useful in construction sites. It is also useful in a warzone. Their designs reflect their purposes. Of course, because their range of motion needs to be optimised for their functions. Outside of their more sinister sides, they can look cute. Spot is built to walk in a rough environment, it uses its sensors in order to gain more information of this environment, and the more it walks, the more data it can collect, which in turn leads it to become more stable. However, it is also given a sweet name and a fun colour, so of course there can be no bad intention behind its motion, right?

Now what?

During the writing of this paper, I kept wishing I could have gone into Marshall McLuhan’s influential The Medium is the Message, with a conclusion that said something like “the movement is the message”. And yet, while there was no fitting moment to do so, I think this is still a sound conclusion. The movements of the robot, whichever robot in question, tells you something about how it was made, and for what purposes. Both Spot and the Roomba, no matter how different their specific physical affordances, can be seen as lively characters with the range of motion they do have. There is no intention to look human, which is what makes them look alive. They can convince their viewers that they are what they seem, all the while being able to be used for less savoury situations. Their movements suggest their intentions.

So, what does its movement imply? What can these machines tell us about what they do? Not much, on its own. It is rather the environment that gives meaning to it. Spot is highly versatile, capable of many different movements, useful for scouting rough terrain and executing dance moves. Its movements can be programmed, but the way it is used can only be decided by the user. The Roomba can map your floor, but it can also just suck up dust while occasionally bumping into furniture where it did not expect to find it.

There is no concrete advice here. There is no guide on how to best animate your robot. Does your robot have arms? Use them. If you don’t need them, you might as well rip them off. Is your robot fast? Is it slow? Lean into that. If you are on the other side of this, if you are building the hardware, think about what will benefit the final robot’s ability to communicate. Find a way to make the robot seem alive and there is a good chance people will be endeared by it. The most important part is finding a way to move the robot in a way that makes sense for its physical allowances, and the environment they will be placed in, the rest is up to the user.

[1] Peter Eckersall and Kris Verdonck, “On Composite Bodies and New Media Dramaturgy,” in Performance and Posthumanism: Staging Prototypes of Composite Bodies, ed. Christel Stalpaert, Kristof van Baarle and Laura Karreman (Cham: Springer International Publishing, 2021), 111.

[2] Maaike Bleeker and Marco C. Rozendaal, “Dramaturgy for Devices: Theatre as Perspective on the Design of Smart Objects,” in Designing Smart Objects in Everyday Life: Intelligences, Agencies, Ecologies, ed. Marco C. Rozendaal, Betti Marenko and William Odom (New York: Bloomsbury, 2021), 50.

[3] Susan Leigh Foster, Choreographing Empathy: Kinesthesia in Performance (London and New York: Routledge, 2011), 6-7.

[4] Susan Leigh Foster, “Movement’s contagion: the kinesthetic impact of performance,” in The Cambridge Companion to Performance Studies, ed. Tracy Davis (Cambridge: Cambridge University Press, 2009), 48-49.

[5] Foster, “Movement’s contagion,” 51.

[6] Académie militaire de Saint-Cyr Coëtquidan, Twitter post, April 6, 2021 (5:35 pm), accessed January 26, 2022. https://twitter.com/SaintCyrCoet/status/1379457690020294665?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1379457690020294665%7Ctwgr%5E%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fwww.boston.com%2Fwp-admin%2Fpost-new.php

[7] 11 Network Australia, “The Evolution of Boston Dynamics,” Youtube video, accessed January 15, 2022, https://www.youtube.com/watch?v=D_Vc_yDvU24.

[8] Cristina Van Orden, “iRobot Military Discount,” Military Discount Shop, accessed January 26, 2022, https://militarydiscounts.shop/irobot-military-discount/.

[9] Elizabeth Ann Jochum and Todd Murphey, “Programming Play: Puppets, Robots, and Engineering,” in The Routledge Companion to Puppetry and Material Performance, ed. Dassia N. Posner, Claudia Orenstein and John Bell (London: Taylor and Francis Group, 2014), 310.

[10] Jochum and Murphey, “Programming Play,” 310.

[11] “Sophia,” Hanson Robotics, accessed January 25, 2022, https://www.hansonrobotics.com/sophia/.

[12] Jochum and Murphey, “Programming Play,” 310.

[13] Mari Velonaki and David Rye, “Designing Robots Creatively,” in Robots and Art: Exploring an Unlikely Symbiosis, ed. Damith Herath, Christian Kroos and Stelarc (Singapore: Springer, 2016), 382-3.

[14] Velonaki and Rye, “Designing Robots Creatively,” 391.

[15] Velonaki and Rye, “Designing Robots Creatively,” 380.

Bibliography

Bleeker, Maaike and Marco C. Rozendaal. “Dramaturgy for Devices: Theatre as Perspective on the Design of Smart Objects.” In Designing Smart Objects in Everyday Life: Intelligences, Agencies, Ecologies, edited by Marco C. Rozendaal, Betti Marenko and William Odom, 43-56. New York: Bloomsbury, 2021.

Eckersall, Peter and Kris Verdonck. “On Composite Bodies and New Media Dramaturgy.” In Performance and Posthumanism: Staging Prototypes of Composite Bodies, edited by Christel Stalpaert, Kristof van Baarle and Laura Karreman, 111-124. Cham: Springer International Publishing, 2021.

Foster, Susan Leigh. Choreographing Empathy: Kinesthesia in Performance. London: Routledge, 2011.

Foster, Susan Leigh. “Movement’s contagion: the kinesthetic impact of performance.” In The Cambridge Companion to Performance Studies, edited by Tracy Davis, 46-59. Cambridge: Cambridge University Press, 2009.

Hanson Robotics, “Sophia,” Hanson Robotics. Accessed January 25, 2022. https://www.hansonrobotics.com/sophia/.

Jochum, Elizabeth Ann and Todd Murphey. “Programming Play: Puppets, Robots, and Engineering.” In The Routledge Companion to Puppetry and Material Performance, edited by Dassia N. Posner, Claudia Orenstein and John Bell, 308-321. London: Routledge, 2014.

Van Orden, Christina. “iRobot Military Discount.” Military Discount Shop. Accessed January 26, 2022. https://militarydiscounts.shop/irobot-military-discount/.

Velonaki, Mari and David Rye. “Designing Robots Creatively.” In Robots and Art: Exploring an Unlikely Symbiosis, edited by Damith Herath, Christian Kroos and Stelarc, 379-401. Singapore: Springer, 2016.