Human-Robot Interaction

A new wave of humanoids entered the scene, and their creators promised us a bright future. Atlas, Figure, and Optimus are intended to work in spaces that are designed for humans. They are not only targeted at factories, but also at our homes and families. But what promises can they actually meet? Dwain Allan and I interviewed Will Jackson and Robert Riener on the future of humanoids.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

Artificial Intelligence promise to improve many aspects of our lives. The current hype, in particular around Large Language Model, raises the questions what of these bold visions of the future is science and what is fiction. AI is not new in HRI. For many decades researchers used it to enable robots to better interact with humans. In this episode I talked with Julie Adams and David Kaber about what AI can really do for HRI.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

The peer review process is essential to modern science. Researchers conduct studies and submit their results to a journal. An editor manages a review process involving external experts. But what happens when you study the peer review process itself? How do scientific organisations react when they become the subject of an experiment? Not well, to say the least.

This live recording is from the Nerd Night at the Little Andromeda Theatre in Christchuch, New Zealand.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

Should you use chatGPT for podcasting? I asked chatGPT this question and used Descript’s Overdub feature to voice the response. Generating this episode was quicker than any other I ever created. Is it any good? Probably not. Still, it shows how quickly it is possible to generate superficial content. This might be useful for enabling robots to have superficial conversations with people. Just don’t expect any originality or accuracy.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

Publishing your human-robot interaction study in a journal is an excellent way to share your insights. But in which journal should you publish and what do the journals expect? In this episode, we talk to editors from the three dedicated HRI journals, Agnieszka Wykowska (International Journal of Social Robotics), Selma Sabanovic (ACM Transactions on Human-Robot Interaction) and Bilge Mutlu (Frontiers in Robotics and AI | Human-Robot Interaction). We talk about Open Access publishing and what the future of scientific publishing might look like. Besides the three dedicated journals, there are also some journals that encourage HRI topics without focusing completely on them. I talked with Kerstin Dautenhahn from the Interaction Studies journal and Ramanarayan Vasudevan from the IEEE Transactions on Robotics.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

Here is a table with all the key indicators for the journals. The full table is available online.

Softbank Robotics sold their Nao and Pepper robots to the United Robotics Group (URG) which reversed the name of this business back to its orginal “Aldebaran”. In this episode Dwain Allan and I discuss the uncertain future of Nao and Pepper based on direct correspondence with URG. We try to answer the question whether you should still invest in this robotic platform. Is Aldebaran another zombie robotic company?

The transcript of the episode is available as a PDF.

• Cobiotx at URG
• SoftBank shrinks robotics business, stops Pepper production
• Repair of broken Nao robot battery pack

Are robots zombies? This might seem like a strange question at first, but it leads to one of the most important questions in science today: what is consciousness and can robot’s become conscious? These questions fascinate many people. And when Blake Lemoine suggested that Google’s latest AI Lambda had become sentient, it triggered a worldwide media frenzy. In this HRI podcast episode, I talk with Professor Jack Copeland about machine consciousness.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript. 

• Is LaMDA Sentient? — an Interview
• Artificial Intelligence: A Philosophical Introduction by Jack Copeland

Autonomous Vehicles (AV) are the riskiest form of human-robot interaction. One the one hand they offer unparalleled improvements to the safety and comfort of drivers, passengers and other traffic participants. They also promise to reduce emission. On the other hand, they demand new considerations for trust and responsibilities in human-robot interaction. The field of tension between autonomy, trust and liability can only be manoeuvred on the basis of objective data. Government agencies will have to implement clear obligations and standards for reporting the performances of AVs. This is the only way for consumers to make informed decisions about their purchases and the associated risks. In this podcast episode we will critically review the available data, the market situation and the recent developments.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript. A full paper with references is available.

• Tesla Deaths
• San Francisco Police Pull Over Driverless Car
• We Drove a Tesla with No One in the Driver’s Seat. That’s Not a Good Thing

In the last HRI podcast episode we talked about the role of emotions for humans and robots. The interview with Douglas Campbell truly inspired me and there was not enough time in the last episode to go into the full depth of our conversation. So here it is, our full discussion about emotions and consciousness in robots.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

• Problems of Philosophy – The Limits of Inquiry by C.McGinn
• Other Minds by Stanford Encyclopedia of Philosophy
• Artificial consciousness by Wikipedia

What roles do emotions, affect and moods play for humans and robots? We will discuss how robots can sense, reason and express emotions. Furthermore, we talk about if emotions make robots more human. I had the pleasure to talk with Rosalind Picard (MIT), Lola Canamero (CY Cergy Paris University), Hiroshi Ishiguro (Osaka University) and Douglas Campbell (University of Canterbury) about these exciting questions.

The transcript of the episode is available as a PDF. You can also follow the episode with subtitles through Descript.

• Facial Action Coding System – Paul Ekman Group
• Affective Computing by Rosalind Picard
• Geminoid Robots