Antropo: An open-source platform to increase the anthropomorphism of the Franka Emika collaborative robot arm

Robot-to-human communication is important for mutual understanding during human-robot collaboration. Most of the current collaborative robots (cobots) are designed with low levels of anthropomorphism. Therefore, the ability of cobots to express human-like communication is limited. In this work, we present an open-source platform named Antropo to increase the level of anthropomorphism of Franka Emika—a widely used collaborative robot arm. The Antropo platform includes three modules: a camera module for expressing eye gaze, a light module for visual feedback, and a sound module for acoustic feedback. These modules can be rapidly prototyped through 3D printers, laser-cutters, and off-the-shelf components available at a low cost. The Antropo platform can be easily installed on the Franka Emika robot. The added communication channels can be synchronised with the robot’s motions to enhance mutual understanding. All hardware CAD design files and software files are released. The platform can be used to study human-like behaviours of cobots and the effects of these behaviours on different aspects of human-robot collaboration. We demonstrate the Antropo platform in an assembly task in which the Franka Emika robot expresses various human-like communicative behaviours via the added communication channels. We also present two industrial applications in which the Antropo platform was customised for the Universal Robots UR16e.

For cobots to express social cues, they should possess a higher degree of anthropomorphism, human-like qualities, either in their appearance or behaviour, or both (Duffy 2003).Anthropomorphism is attributing human traits, emotions, or intentions to non-human entities (Shin & Kim 2020).It has been shown that the addition of anthropomorphic traits to robots increases the sympathy and willingness to work and interact with them (Oistad et al. 2016).Previous studies also have shown that robots with higher levels of anthropomorphism can increase people's perception of the robot's animacy, likability, intelligence, and enjoyment, as well as their intention to use the robot and their engagement with the task (Sauer et al. 2021, Fischer et al. 2015, Faibish et al. 2022).
The research aim has been formulated as follows.
In this work, we propose Antropo, an open-source platform that enhances the anthropomorphism of the widely used Franka Emika collaborative robot arm, enabling it to express a more diverse range of social cues.
Comment: Materials: I wonder whether the placement of the features is not limiting the robot's task performance or whether the system will not get damaged too easily as the gripper is the most used part of this robot.Perhaps it would be good to address this.Furthermore, the choices for specific colors to indicate specific actions might not be as intuitive because certain colors also have an emotional connotation e.g.red being -angry and blue -sad.There is a paper by Johnson, D.O., Cuijpers, R.H., van der Pol, D.: Imitating human emotions with artificial facial expressions.International Journal of Social Robotics 5, 503-513 (2013) That looked at the meaning of different colored eyes for robots and how they are perceived.Perhaps that can help you to strengthen and or motivate your choices for colors of the eyes Response: Thank you for pointing out these issues.First, we added two industrial applications using the Antropo platform with different movements, which demonstrates that it does not limit task performance.
Two customised versions of the Antropo platform have been developed for the Universal Robots UR16s in two distinct industrial applications: gear deburring and assembly of an electric cabinet, see Fig. 5.In both cases, the robot performed different movements, and the Antropo platform did not restrict any of the robot's movements.

Gear deburring
Assembly of an electric cabinet Second, we consider Antropo being damaged a low risk because the light module is made with a hard plastic shell that protects the electronics inside.In case the shell is broken, it can also be replaced easily without influencing other parts of the platform.We have added this point to the Discussion.
Since Antropo is integrated into a collaborative robot, safety considerations are essential when operating the Antropo platform.Antropo was specifically designed to ensure that it does not cause any movement limitations of the Franka Emika or result in self-collisions of the cobot and the platform.Notably, our implementation on the UR16 showcases the platform's capability to facilitate the manipulation of substantial objects without impeding the robot's kinematic performance.Nevertheless, it is practical to validate the programmed robot trajectory before execution to avoid unexpected collisions, particularly when the platform is customised with another Franka's end-effector or other collaborative robot arms such as different versions of the Universal Robots.Regarding external collision, the light module's resilient plastic shell shields the internal electronics, thereby minimising the probability of damage.In the event of shell breakage, the module can be easily replaced, preserving the platform's functionality while maintaining the integrity of other system components.
Third, our light colour selection is based on the colour codes defined in the international standard IEC 60073:2002 for human-machine interaction colour codes defined in IEC 60073:2002.These colours in this standard are also used for traffic lights and signs in road traffic.However, we agree that with a higher level of anthropomorphism, the colours can be linked to emotions.In the scope of this open-source design paper, we did not address this issue but in future studies.We have added this point to the Discussion.
The selection of light colours for the Antropo platform was carefully considered and based on international standards that are commonly used in human-machine interaction and light traffic.These standards were appropriate for the Antropo platform as they provide clear and easily distinguishable colours that are suitable for conveying information to humans.However, as the level of anthropomorphism increases, the colours used in combination with the robot's embodiment can begin to take on additional meaning beyond their intended purpose.As can be seen in previous studies on social robotics, the use of certain colours in combination with the robot's behaviour can elicit emotions and potentially leads to emotional responses from humans (H äring et al. 2011, Johnson et al. 2013, De Beir et al. 2016, Guo et al. 2019).Future research should aim to investigate this issue with the Antropo platform.
Comment: Results: it would be nice to implement some user evaluations of your setup to see how it is perceived to add more weight to the accuracy of the proposed actions.Even if it is just something very short as some interview of an exploratory survey.

Response:
We have included two demonstrations from our ongoing projects.In the first demonstration, in an online study, we showed that Franka with the Antropo platform was rated higher in user perception and acceptance compared to no social cues (Cao et al. 2023).
In the second demonstration, Antropo was customised for the Universal Robots UR16e.Preliminary interview data showed that factory operators agreed that the Antropo platform helped them understand what the robot is doing, and they are willing to work with the robot with Antropo.

Reviewer #2
Comment: This paper focuses on robot-to-human communication for human-robot collaboration.The paper indicates that currently there are a few collaborative robots (so called cobots) that are designed with anthropomorphic features.Further, the paper presents an open-source platform called Antropo that aims at increasing the anthropomorphism of Franka Emika cobot platform, through, for instance, added communication channels.The Antropo platform includes a camera module expressine eye gaze, a light module for visual feedback, and a sound module.
I will try to give the authors some input that can help them improve their paper.Please, see my comments below.
Response: Thank you for your valuable feedback on our work.We have carefully considered your comments and have made revisions accordingly.
Comment: I would suggest that the title is adjusted to: "Antropo: An open-source platform to increase the anthropomorphism of the Franka Emika collaborative robotic arm".In addition, that Franka Emika is a collaborative robotic arm should already be specified in the abstract.

Response:
We have adjusted to title and the abstract based on your comment to specify Franka Emika collaborative robotic arm.New title: "Antropo: An open-source platform to increase the anthropomorphism of the Franka Emika collaborative robot arm".
In this work, we present an open-source platform named Antropo to increase the level of anthropomorphism of Franka Emika -a widely used collaborative robot arm.
Comment: In addition, collaborative robots/cobots should be clearly defined.Note, that the term "collaborative" is a positive value-ladden word that in some fields (e.g., Computer Supported Cooperative Work -see the work of Kjell Schmidt and Liam Bannon on collaboration vs. cooperation) indicates that a goal is achieved without conflicts, whereas the term "cooperation" (as opposed to "collaboration") is a more neutral term, which does not imply that the goal is achieved in a friction-free way.To avoid confusion, indicate the definition of "collaborative robots" that you follow, and the corresponding reference.

Response:
We agree that in other fields, the term "collaborative" might cause confusion.To avoid possible confusion suggested by the reviewer, we added a definition of "collaborative robot" according to the ISO 8373:2012 in the field of collaborative robotics.
Collaborative robots (cobots) are designed to work in close proximity collaboratively with human workers.In the ISO 8373:2012 standard, they are defined as robots designed for direct interaction with a human (ISO 2012).
Comment: Besides the "positive effects" of including a cobot in industry and/or in everyday life, please also specify eventual disadvantages (see the work from Martinetti et al. on safety, including the discussion on psychological safety: https://www.frontiersin.org/articles/10.3389/fceng.2021.666237/full).

Response:
We have added the disadvantages of cobot deployments mentioned in Martinetti et al. (2021), which leads to the importance of mutual understanding.This is what we would like to achieve through the Antropo open-source platform.
While collaborative robots have certain advantages in the workplace, they might pose physical risks and cause psychological stress (Martinetti et al. 2021).To ensure safe and effective collaboration, it is important to establish mutual understanding between humans and cobots to ensure safe, seamless and productive human-robot interaction (Sciutti et al. 2018).
Comment: Further, the authors argue for the anthropomorphization of these robotic arms: " In this work, we aim to increase the level of anthropomorphism of the Franka Emika robot -a commonly used robot platform across research labs and seen as a reference platform for academic research due to its low-level access to robot control, and comparatively lower cost [13,14]." Please, bring more arguments on why these robots should be anthropomorphized.Be aware also of the critique from Johanna Seibt on the use of term "anthropomorphism" associated to robots.She and her colleagues argue rather for the use of term "sociomorphism", where sociomoprhism can be seen at different levels.Please, check out her work: https://pure.au.dk/portal/en/publications/ sociomorphing-not-anthropomorphizing(837953c2-3dc4-4507-9803-a3a3b6be3e8e).html.
Response: Thank you for providing a very interesting paper and relevant to our work.We used "anthropomorphism" because it is a widely used concept in the field.We added our recommendation of sociomoprhism vs anthropomorphism to improve the Introduction part of our paper.
It should be noted that we adopted the common assumption in the social robotics field that social interactions with robots are due to anthropomorphism while sociomorphism, the perception of actual non-human social capacities, can also play a role (Seibt et al. 2020).
Comment: Besides specifying the aim and bringing more arguments that support this aim, please, clearly specify the research question.
Response: Since our work is a open-source design paper, we formulate the research aim as follows.
In this work, we propose Antropo, an open-source platform that enhances the anthropomorphic capabilities of the widely used Franka Emika collaborative robot arm, enabling it to express a more diverse range of social cues.
We demonstrate how we reached the aim by the preliminary results of two experiments.
To demonstrate the platform's capabilities, we present two experiments conducted under EU-funded projects.In the first experiment, we used the original Antropo platform with the Franka robot to investigate the effects of multi-modal social cues on human-robot collaboration.In the second experiment, we used a customised version of the platform with the Universal Robots UR16e to explore the performance of a shared task between a human and a cobot in an industrial setting.
Comment: Regarding the method chosen, please, describe a scientific method (e.g., prototyping) rather than only describing the procedure and materials.

Response:
We have added the standard engineering design process in the Materials and Methods, which we followed during the design process.
When designing the Antropo prototype, we followed the standard engineering design process including four steps and the two key features of open-source hardware: the use of off-the-shelf components and the online availability of designs and materials (Maia Chagas 2018, Bonvoisin et al. 2017).