Category: Future of work

The future hybrid meetings: Now you can “whisper” to your virtual colleague

Post author By Susanne Brøndberg
Post date August 20, 2024

20 August 2024

The future hybrid meetings: Now you can "whisper" to your virtual colleague

Researchers have developed a new technology that enables participants in hybrid meetings to communicate more effectively. This innovation makes it easier for online attendees to build strong connections with colleagues in a world where hybrid meetings have rapidly become the norm.

Picture this: It’s Monday, and the workweek kicks off with a team meeting. Some colleagues are physically present, while others join from home — or perhaps even from a dentist’s waiting room. It’s a typical hybrid meeting.

From your virtual seat, you overhear soft exchanges among the participants in the physical room:

“Did you have a nice weekend?”

“What did she mean by that?”

“How did Friday’s meeting go?”

As an online participant, the challenge is that you cannot join in these side conversations. It can also be difficult to catch the subtle social interactions that occur during meetings – interactions that are essential for building better relationships with colleagues. But this does not have to be the case anymore.

With the program WhisperChannel, researchers have developed a tool that allows all participants in hybrid meetings to “whisper” to each other online. This technology holds the potential to significantly improve the quality of hybrid meetings.

A walkie-talkie for hybrid meetings

Ph.D. student Qianqian Mu from the Department of Computer Science at Aarhus University is the driving force behind WhisperChannel, which she developed in collaboration with experts from the Alexandra Institute. This web app allows all participants in a hybrid meeting to send audio messages to selected individuals, much like a walkie-talkie designed for the hybrid workforce.

“Most online meeting systems only support communication with the entire group. There is a missing backchannel that is more intimate, enabling participants to feel closer, even when they are physically distant,” says Qianqian Mu. She adds, “We created WhisperChannel to help meeting participants build stronger relationships and to even out the imbalance between them. We’re already seeing that both online and in-person attendees feel more connected”.

WhisperChannel was developed through testing, interviews, surveys, and observations in companies such as Arla, BEC, Bankdata, EventSpace, LTIMindtree, and Unlikly. Additional features will be added to WhisperChannel in the future, which will further strengthen colleague connections.

“We are currently experimenting with the ability to send vibrations to each other. This feature has been very popular because it simulates a physical touch. Looking ahead, I believe there is potential to integrate this technology into wearables like watches and glasses,” says Qianqian Mu, who has published a scientific paper on the need for WhisperChannel.

A collaboration between companies and researchers

WhisperChannel is part of the broader REWORK project, financially supported by the Digital Research Centre Denmark (DIREC). Researchers have collaborated with the aforementioned Danish companies, which have provided valuable insights.

Several corporate partners are currently testing the technology, and researchers are witnessing significant interest from the business sector, which is eager to optimise the new hybrid work environment.

“We can never make hybrid meetings identical to in-person ones. Instead, we must embrace the differences and work with the inherent asymmetry of hybrid meetings,” says REWORK project leader Eve Hoggan, Associate Professor at the Department of Computer Science at Aarhus University.

Since 2022, Eve Hoggan has brought together researchers, companies, cultural institutions, and artists to rethink hybrid work, which gained significant popularity after the COVID-19 pandemic.

Companies are keen to improve the meeting experience or contribute specific technologies. The researchers from various research domains offer diverse perspectives on the overarching challenges. The cultural sector can help communicate these initiatives to a broader audience. With WhisperChannel, many hours of cross-disciplinary collaboration have resulted in a tangible tool with global potential.

“DIREC has enabled us to assemble a group that can address the challenges of hybrid work from multiple angles. This has allowed us to share insights across different fields and develop concrete tools that can shape the future of hybrid work,” concludes Eve Hoggan.

The REWORK project is set to conclude in 2025.

Learn more about REWORK here.

Tags p18

Future of work News

Error source or necessity? Researchers explore the human role in future production

Post author By Susanne Brøndberg
Post date July 2, 2024

2 July 2024

Error source or necessity? Researchers explore the human role in future production

The interplay between humans and technology is a subject of intense debate. Is human involvement necessary if technology can handle the work on its own? A research collaboration between Copenhagen Business School and the University of Southern Denmark, funded by DIREC, highlights the importance of human involvement in automated systems.

As technology becomes an increasingly significant part of industrial processes and daily life, understanding the foundations of human-machine collaboration is becoming increasingly important.

In the research project Cyber-physical systems with humans in the loop, supported by the Digital Research Centre Denmark (DIREC), researchers from the Department of Digitalization at Copenhagen Business School and the Software Engineering Section at the Maersk Mc-Kinney Moller Institute at the University of Southern Denmark have examined the significance of the human role in cyber-physical systems (CPS).

CPS are systems that integrate physical objects with software enabling these components to interact and exchange information, for example in advanced industrial processes.

The researchers aim to establish a better conceptual framework and a foster a shared understanding of human collaboration with technology—a shared understanding that is currently lacking.

“Many in the industry view humans as a source of error and risk factors. They wish to automate humans out of the equation, excluding them from system design. However, much of the academic literature sees this perspective as problematic. Instead, systems should be designed to incorporate human roles. This creates a disconnect between academic theory and industry practice,” says Torkil Clemmensen, Professor at the Department of Digitalization at Copenhagen Business School.

Collaboration between humanities and engineering

The research team has combined humanistic and technical approaches in their project, which includes literature reviews, software development, and interviews with selected industry partners.

Combining different research domains is a key success factor for the project and has brought many valuable perspectives, according to Mahyar Tourchi Moghaddam, Associate Professor at the Maersk Mc-Kinney Moller Institute at the University of Southern Denmark. With his technical background, he has developed an artificial intelligence that ensures humans remain “in the loop” when machines make decisions.

“We are investigating the subject from various perspectives, and it has been very beneficial not always to think in terms of technical solutions but to consider how systems can best involve humans,” he says.

Humans play an essential role

The collaboration between the two research domains has led the researchers to conclude that humans play an essential role in working with machines. In fact, humans are difficult to eliminate when cyber-physical systems need to operate optimally. The challenge now is to identify the functions where humans add the most value, which should be the focus of future research.

“We need to find the balance between technological innovation and human capabilities. Therefore, more concrete interdisciplinary research is necessary to promote a harmonious coexistence between humans and machines,” concludes Torkil Clemmensen.

The conclusions from “Cyber-physical systems with humans in the loop” will be published later in a scientific article.

Swarms of robots are being deployed on the fields – What does it take to expand the use of them?

Post author By Susanne Brøndberg
Post date November 21, 2023

21 November 2023

Swarms of robots are being deployed on the fields – What does it take to expand the use of them?

Danish farmers are ready to embrace new technologies to support the green transition and ensure smarter production. Multi-robot systems are a crucial part of the solution, but barriers need to be dismantled and teething problems eliminated for seamless interaction between farmers and robots.

Self-driving robots are replacing diesel-powered giant machines, and multi-robot systems enable several robots to collaborate in the fields. Precision spraying of fertilizers and pesticides reduces the use of spray chemicals.

There are both environmental and efficiency gains in entrusting fieldwork to robots, and technology plays a vital role in the agricultural green transition.

“One of the major problems in agriculture is that farm machinery is getting larger and larger. However, when large machines traverse the ground, they compact the soil, requiring a significant amount of energy to repair the damage they cause. If instead, we deploy smaller, autonomous robots, we can increase efficiency without causing damage to the environment.” 
– Anders Lyhne Christensen, Professor, University of Southern Denmark, UAS Center

The development imposes new requirements on both technology and users. In the HERD project, funded by DIREC – Digital Research Centre Denmark, Aalborg University’s expertise in designing user interfaces, University of Southern Denmark’s (SDU) knowledge of robotics, and Copenhagen Business School’s (CBS) insights into market creation and business models are combined with use cases from various companies developing robot systems.

AGROINTELLI, a Danish scaleup, is one such company working to break down barriers preventing farmers from adopting new technologies. Alea Scovill, R&D Manager at AGROINTELLI, emphasizes the importance of addressing factors like price, robustness, and user-friendliness to facilitate wider robot adoption.

“At AGROINTELLI, efforts are being directed towards breaking down some of the barriers currently preventing several farmers from adopting new technologies – a challenge encountered by the majority of field robot companies in the EU, says R&D Manager Alea Scovill from AGROINTELLI.

“If farmers cannot see how the robot fits into the farm and can be used without significant instruction, sales are lost. Price, robustness, and user-friendliness are other parameters influencing the adoption and serving as barriers for more farmers to embrace the robots,” explains Alea Scovill, who is in close dialogue with the involved researchers from CBS.

The role of CBS researchers in the project is precisely to explore the market challenges and commercial opportunities in the technology, and what it takes to mature the market. PhD student Alexandra Hettich, for instance, has interviewed various stakeholders such as sales personnel and dealers, and will soon interview the farmers.

“Agriculture is particularly interesting as a domain. With the introduction of robots, the farmer’s work is significantly altered, and the obstacles to a successful implementation of this groundbreaking technology vary in nature. Therefore, we need to analyze the diversity of obstacles before developing concrete solutions to overcome them,” says Alexandra Hettich, PhD student at CBS.”

The collaboration uncovers various aspects of the technology

According to Professor Anders Lyhne Christensen from the University of Southern Denmark, the project leader for the project, the results are particularly interesting because they cover all aspects of technology, focusing on both the technological challenges, user experience, and the commercial aspects of agricultural robots as a business area.

“At SDU, we work with multi-robot systems and focus on how to make robots do what they need to do and provide the user with the information they need. Aalborg University works on user interfaces and user interfaces, such as how users can keep track of what robots have done, what they are currently doing, and what they will do in the future. In other words, how to give the user the right buttons to turn. Finally, CBS focuses on the business part for companies developing these robots and what business models may be promising for them. How can they access the market, and what happens at the other end with the organizations that need to use multi-robot systems, how do they change?” explains Anders Lyhne Christensen.

The focus is largely on users’ understanding and use of the technology, he elaborates.

“We can certainly create robots and program them to do this and that, but getting them to work in the real world requires that people can control them. What we are working on is therefore the interaction between the AI in the robots, the people who have to control the robots, and the organizations around them.”

“It doesn’t work if the farmer has to keep an eye on the robot while it performs the task – not much is gained then. Instead, it is important to be able to oversee what several robots are doing at once.”

It may also be that the farmer himself does not have to monitor the robots, but a company that monitors robots for 50 farmers at a time, and then it changes those organizations because new job functions and responsibilities come with the technology.”

Alea Scovill is pleased with the collaboration with the researchers. It works well, says the R&D manager.

“The flow of information between the partners in the project has been relatively smooth. At AGROINTELLI, we have primarily worked with CBS and Aalborg University because their research areas fit well with our situation. CBS is investigating the market obstacles for ROBOTTI. And at Aalborg University, the researchers have developed a new proposal for a user interface for remote monitoring of multiple robots, and they will soon interview an agricultural school about the experience of the new user interface,” says Alea Scovill.

ABOUT THE HERD PROJECT

In the HERD project, researchers, along with industrial partners, aim to develop technologies that enable end-users to engage and control systems consisting of multiple robots. The goal is to enhance the value of industrial products by enabling faster and more cost-effective completion of current tasks and addressing entirely new tasks that require coordinated robot efforts.

Project period: 2021 to 2025
Budget: DKK 17.08 million
Partners: University of Southern Denmark, Aalborg University, Copenhagen Business School, AGROINTELLI, ROBOTTO, and the Danish Technological Institute.

More about the project

Tags P17

Future of work News Researcher spotlight

For four decades, Susanne has been shaping the digital everyday life of Danes

Post author By Susanne Brøndberg
Post date October 4, 2023

4 October 2023

For four decades, Susanne has been shaping the digital everyday life of Danes

Susanne Bødker’s research in Human-Computer Interaction (HCI) has had a profound influence for decades on how we interact with technology in our daily lives and work. In September, she celebrated her 40th anniversary at Aarhus University.

Photo: Morten Koldby

IT solutions should be designed by humans for humans; digital tools should make a difference in everyday life and function like an extended arm, seamlessly integrated without us having to consciously think about it.

The research field of Human-Computer Interaction (HCI) plays a central role in the technology that surrounds us daily, and in the way the job market is evolving, ensuring that new IT solutions effectively support human use.

One of Denmark’s leading researchers in the HCI field is Susanne Bødker, a computer science professor at Aarhus University, honored multiple times for her research results in human-machine interaction.

Since the 1980s, this researcher from Aarhus has been involved in designing the digital everyday life and work of the Danes, ensuring that technology develops in a constructive way and critically addresses challenges and opportunities.

Currently, she is particularly focused on how hybrid work challenges companies and employees, and how it fundamentally alters the interaction and relationship between people in a workplace, for better or worse.

– Hybrid work is only becoming more prevalent, so we need to critically consider the possibilities and limitations of technology, as well as the way we organise and lead. When a workplace with several hundred employees, for example, decides that all activities should be hybrid going forward, it imposes new demands on personnel management. It changes the very nature of work and meetings when employees must always be able to participate remotely. This affects what can be shared, when and how – it essentially changes everything participants see, hear, and experience because on the screen, we are still only ‘flat people,’ says Susanne Bødker.

Examine your organisation critically and inquisitively

When advising organisations on how to adapt to being a modern hybrid workplace, this entails considering technology, physical environments, and the managerial aspects of hybrid work.

– Companies face vastly different challenges, and the technology must be integrated into the specific context. Are you a software company with employees all over the world, struggling with the issue that people are reluctant to move to Aarhus? Are you a bank looking to replace physical customer meetings with online ones? Do you simply want people to have the freedom to work from home and only physically come into the office a few days a week? In that case, it is necessary to organise differently so people come into the office on the same days. Every company needs to address its own reality and current challenges.

Her extensive research in user interfaces and user experiences has led to new methods and theories that have gained international attention. In 2017, she received an ERC Advanced Grant of over 2 million euros from the European Research Council for research in user interfaces for complex human use of computers and the research project “Common Interactive Objects.” The goal was to explore the possibility of building open and shareable platforms and communities based on the user’s – not the computer systems’ – terms.

Most recently, she is participating in the REWORK project, funded by the Digital Research Centre Denmark. REWORK is a multidisciplinary research project where researchers, various companies, and three recognised artists explore the future of the hybrid workplace particularly focusing on new technologies that support aspects such as human needs, relational and articulation work, as well as embodiment and presence.

Tags Impact story, p18

Future of work News

DIREC Interview: Professor Henrik Christensen on developments in robot technology

Post author By Susanne Brøndberg
Post date October 3, 2023

3 October 2023

DIREC Interview: Professor Henrik Christensen on developments in robot technology

Professor and director of the Contextual Robotics Institute, Henrik Christensen from UC San Diego, recently visited Denmark. We met the world-renowned robotics expert for a discussion on Danish tech research compared to American and Swedish research and on the next major advancements in robotics technology.

He began his career as a researcher at Aalborg University and later moved to Stockholm as head of the Center for Autonomous Systems at KTH Royal Institute of Technology.

His career led him to the United States, where Henrik Christensen has held several prominent roles at world-leading universities. Today, he is a researcher and head of the Contextual Robotics Institute at the University of California San Diego. Additionally, he is an entrepreneur and investor.

Recently, he visited Aarhus for the Robot Festival ‘Wild Robots,’ where he was invited as a keynote speaker. We caught up with him to discuss the development of robotics technology and the prospects of using robot technology in areas such as healthcare, the green transition, and more.

First, we asked him about the differences that characterize research environments in Denmark, Sweden, and the USA and how Denmark can strengthen its position on the international research scene.

What is the biggest difference between conducting tech research in Denmark and abroad, and what can we learn from the USA and Sweden?

“In the USA, all research is extremely results-focused. You have to be sharp on the impact of your research project and ensure that you publish the results in the most reputable journals to maximize its impact. Focus is constantly on how your project can make a difference and where your research can have the most significant impact. Mainstream research is uninteresting in the USA.”

In Sweden, there is a long tradition of academic research environments, and it is largely one university that sets the agenda for tech research, he explains.

“The Royal Academy of Science dominates tech research in Sweden. In Denmark, we have a tradition where each university individually defines its priorities, with less focus on the big picture. In Denmark, we are good at ensuring that we conduct research relevant to the industry in the long term, but we are less focused on the long-term strategic impact of tech research.”

Is it about thinking bigger when you have a large budget and is, for example, a university like Berkeley? Does it matter for research whether you have a large or small research budget?

“Of course. I am co-director of a large AI and automation institute with a budget of $25 million. There are not many places in Denmark with a budget of that size, and that is just one of 25 centers in the USA, so we are talking about a lot of money, and the research budget is even larger if we include the defense industry.”

“My impression is that in Denmark, the overall research strategy is less clear. In Sweden, you have companies like Saab, Volvo, Ericsson, and all the pharmaceutical companies that handle industrial research, and then you leave the long-term strategic research to the universities. Companies support universities financially, but the requirement is that universities think big and long-term. My sense is that Denmark does not have the same strategic focus as the USA and Sweden. We have Novo Nordisk and LEGO and a few other heavyweight companies, but compared to the USA, there are not many. Therefore, there is not the same pressure from the industry to influence the overall research strategy or to be involved in deciding who does what.”

As a professor and head of the Contextual Robotics Institute, Henrik Christensen is at the forefront of robotics research, including the development of autonomous robots that integrate advanced artificial intelligence and sensors, used in industries, healthcare, and society as a whole.

From your perspective, what are robot researchers currently most focused on, and what major breakthroughs have propelled research forward?

“We are seeing significant advancements in materials research and nanotechnology. Now it is possible to create a hand that can feel using sensors, and we can integrate 3D printing and materials. This has enabled us to create much more embodied robots than we ever imagined we could.”
“We have made great progress in sensors. If you need to integrate five cameras on a robot, it is not a problem, and at the same time, we have enormous computing power. Sensors, computing power, AI, deep learning, and language-based models have completely revolutionized tech research. We are working with datasets of a size we never imagined.”

What knowledge are we lacking for robots to truly make a difference, for example, in healthcare and eldercare?

“I think it has a lot to do with the fact that we have so far solved the ‘easy problems.’ We can create better pattern algorithms, behavior analysis, and we can predict where people will be in 50 seconds, so if they are seen in the hall, the robot can navigate around them. But the truly constructive, usable user interfaces are more challenging, and we typically leave them to a computer science professor.”

“Computer scientists are good at certain things, but we have weaknesses and blind spots when it comes to the world and people. That is a limitation. We need help to build much more usable models, and that requires involving other disciplines if robots are to become a more common part of everyday life, such as in healthcare and elderly care. We need to involve individuals who have a more practical approach to research and understand cognitive behavior, interaction patterns, and so on.”

We have discussed specific technologies that are advancing rapidly, especially AI and robotics technology. Which trends would you point out as particularly interesting for today’s tech entrepreneurs?

“As a tech entrepreneur, you need to be aware that it takes much longer to work with and secure investments for robotics technology companies compared to companies based on AI. Robotics technology involves hardware and typically takes a long time to develop. If I were to point out particularly interesting areas in robotics technology, it would be robotically controlled limbs and cyber-physical systems – that is, complex systems that integrate computers, networks, and physical components to monitor and control physical processes.”

“Geographically, it is more challenging to secure venture capital in Europe than in the USA. In the USA, there are hundreds of venture firms ready with risk capital, while the approach in Europe is more cautious. Initiatives are underway to ensure better venture capital in Europe, and within a few years, we will see more multinational investment firms turning their attention to Europe because there are untapped opportunities here.”

You can watch the entire interview with Henrik Christensen here.

Tags Impact story

Future of work News

Exhibition gave a twisted glimpse into the future workplace

Post author By Susanne Brøndberg
Post date October 3, 2023

3 October 2023

Exhibition gave a twisted glimpse into the future workplace

The interdisciplinary research project, REWORK, explores the future of hybrid work technologies and experiences. Three Copenhagen-based artists presented their interpretation of how technology in the near future could transform our work lives – directly and indirectly.

Photo: Thinkalike

It was a thought-provoking glimpse into the future when three Copenhagen-based artists recently presented their visions of what our work lives could look like in the future.

Researchers and artists are designing the future hybrid workplace

Post author By Susanne Brøndberg
Post date October 3, 2023

3 October 2023

Researchers and artists are designing the future hybrid workplace

The research project called ‘REWORK’ aims at developing the digital meeting room to make human interaction feel more authentic and personal, much like in a physical meeting.

Photo: Kulturværftet/Thinkalike

The digital meeting room has quickly become a common part of everyday life in many workplaces – a fast, easy, and convenient way to meet, regardless of one’s location.

However, the hybrid workplace has its limitations, and therefore DIREC has decided to fund a research project involving researchers from Aarhus University, University of Copenhagen, IT University of Copenhagen, and Roskilde University to work on developing the future of digital meeting solutions.

Participating in the project are also several large and small companies, and, as something new, three artists have been invited to contribute ideas and visions for the future hybrid workplace. These artists come from the Copenhagen art scene and each offer their perspective on how the digital meeting space can support the relationships and interactions between people, which are limited in today’s meeting solutions.

Seeing the world differently

Kellie Dunn is a PhD in Computer Science at Copenhagen University and has a background in the theater environment. Last autumn, she moved from the USA to Denmark, where she works full time on the DIREC project, titled ‘REWORK – The Future of Hybrid Work.’

“The artists selected for the project already work with the virtual space and, for example, virtual reality. They bring different perspectives on the challenges. How can the future hybrid workplace, for example, account for all the invisible things that happen between people – body language, shared imagery – all the unspoken elements that are part of the physical meeting between people. That is one of the themes that concern the artists,” explains Kellie Dunn.

When more people work remotely, it challenges traditional working methods in companies. For example, nuances are easily lost when we are not physically present together and cannot read each other.

“It’s a significant thing missing in hybrid work compared to the physical workplace, where it is easier for us to read each other’s body language, feel empathy, and see a situation from others’ perspectives,” says Kellie Dunn.

The project illustrates why it is essential to incorporate different perspectives in the development of digital solutions. Typically, developers design solutions that will be widely used, and often developers are men. To ensure that future solutions fit a broad majority of users, it is necessary to challenge the design process, according to Kellie Dunn.

“Collaborating with artists is not very common in research within computer-supported cooperative work (CSCW), whereas within human-computer interaction, similar collaborations have been successful in the past. It is still not common in work-related research, and that’s a shame because it provides entirely new perspectives on the world.

Artists come without a lot of preconceived rules about how things should be, and they are exceptionally good at imagining the future, whereas if you, for example, ask employees in a workplace – they will typically be limited by a practical focus on how their everyday life looks right now.”

The goal is to establish more collaborations across the (tech) industry and the art world to create better products and more productive work systems, she says.

Now we have the opportunity to create the future we want, instead of passively watching as we continue to do things the way we always have.
Kellie Dunn, PhD in Computer Science at the University of Copenhagen

About the REWORK project

Home and hybrid work is here to stay, but what should these work methods look like in the future? Should we simply try to improve what we already have, or can we take a bolder approach and design a different kind of future in the workplace? In collaboration with several companies, this project seeks a future vision that integrates experiences around hybrid work.

The project includes:

Aarhus University
University of Copenhagen
IT University of Copenhagen
RUC
Alexandra Institute
Culture Yard
LTI
EventSPACE
Khora
Unlikly
KeyLoop
Lead
Bankdata
Arla

Explainable AI will disrupt the grain industry and give farmers confidence

Post author By Susanne Brøndberg
Post date July 4, 2023

4 July 2023

Explainable AI will disrupt the grain industry and give farmers confidence

There is a huge potential for AI in the agricultural sector as a large part of food quality assurance is still handled manually. The aim of a research project is to strengthen understanding of and trust in AI and image analysis, which can improve quality assurance, food quality and optimize production.

One of the major critical barriers to using AI and image analysis in the agriculture and food industry is the trust in its effectiveness.

Today, manual visual inspection of grains remains one of the crucial quality assurance procedures throughout the value chain, ensuring the journey of grains from the field to the table and guaranteeing that farmers receive the right price for their crops.

At the Danish-owned family company FOSS, high-tech analytical instruments are developed for the agriculture and food industry, as well as the chemical and pharmaceutical industries.

Since its founding in 1956 by engineer Nils Foss, development and innovation have been high priorities. As a global producer of niche products, staying ahead of competitors is essential.

Hence, collaboration with researchers from the country’s universities is a crucial part of the company’s digital journey. In a project at the National Research Centre for Digital Technologies (DIREC), the company, along with researchers from Technical University of Denmark and University of Copenhagen, aims to map how AI and image analysis can replace the subjective manual inspection of grains with an automated solution based on image processing. The goal is to develop a method using deep learning neural networks to monitor the quality of seeds and grains using multispectral image data. This method has the potential to provide the grain industry with a disruptive tool to ensure quality and optimize the value of agricultural commodities.

The agricultural and food industry is generally a very conservative industry, and building trust in digital technologies is necessary, explains senior researcher Erik Schou Dreier from FOSS. The development of AI, therefore, cannot stand alone. To encourage farmers to adopt the technology, it is crucial to instill confidence in how it works. In this process, researchers use explainable AI to elucidate how the algorithms function.

Today, grain is assessed manually in many places, and replacing manual work with a machine requires trust. Because the work is performed by humans, it is a fairly subjective reference method used today. Humans may not necessarily perform the work the same way every time and can arrive at different results. Therefore, there will be some uncertainty about the outcome.

Mapping and explaining algorithms

– The result is more precise when using AI and image analysis in the process. However, for these new technologies to gain widespread acceptance globally, a model is needed to explain how AI works and arrives at a given result, says Erik Schou Dreier.

Many people have inherent skepticism toward self-driving cars. Self-driving cars need to be even better and safer at driving than us humans before we trust them. Similarly, the AI analysis models we work with must be significantly better than the manual processes they replace for people to trust them. To build that trust, we must first be able to explain how AI analyzes an image and arrives at a given result. That is the goal of the project—to interpret the way AI works, so people can understand how it reads an image.

We typically accept a higher error rate among humans than machines. For us humans to trust the algorithms, they need to be explainable.
Erik Schou Dreier, senior researcher

PhD student Lenka Tetková from Technical University of Denmark is part of the project and spends some days at FOSS’ office. Here, she works with images of grains in two different ways, partly to improve image qualification and partly to better understand how classifications work so they can be enhanced.

– I sometimes use the example of a zebra and a deer to explain how image classification works. Imagine you have a classification that can recognize zebras and deer. Now, you get a new image of an animal with a body like a deer, but the legs resemble those of a zebra. A standard model will not be able to recognize this animal if it hasn’t seen the animal during training. But if you provide it with additional information (metadata) – in this case, a description of all kinds of animals – it will be able to infer that the image corresponds to an okapi, based on its knowledge of zebras, deer, and the description of an okapi. That is, the model will be able to use information not present in the images to achieve better results, explains Lenka Tetková and continues:

– In this project, we want to use metadata about the grains, such as information about the place of origin, weather conditions, pesticide use, and storage conditions, to improve the classification of grains.

Can you find ‘Okapi’ in these pictures? Ph.D. student Lenka Tetková from DTU uses this example to explain how image classification works.

An important competitive advantage

As a global producer of niche products, FOSS must always stay two steps ahead of competitors.

– To ensure there is a market for us in the future, it is crucial to be the first with new solutions. It is challenging to make a profit if there is already a player doing it better, which is why we constantly introduce new digital technologies to improve our analysis tools. And here, collaboration with researchers from the country’s universities is very valuable to us, as we gain new insights and proposed solutions for the further development of our tools, says Erik Schou Dreier and continues:

– In this project, we hope that collaboration with researchers will lead to the development of AI methods and tools that enable us to create new solutions for automated image-based quality assessment and, secondly, that we can increase trust in our product with explainable AI. It is one of the critical themes for us—to create a product that is trusted.

Facts about FOSS

FOSS’ measuring instruments are used everywhere in the agriculture and food industry to quality assure a wide range of raw materials and finished food products.

Traditionally, light wavelengths are measured, and the measurements are used to obtain chemical information about a product. This can include knowledge about protein and moisture content in grains or fat and protein in milk, etc.

FOSS’ customers are large global companies that use FOSS’ products to quality assure and optimize their production—and to ensure the right pricing, so, for example, the farmer gets the right price for their grain.

Deep Learning and Automation of Imaging-based Quality of Seeds and Grains

Project Period: 2020-2024
Budget: DKK 3.91 million

Project participants:

Lenka Tetková
Lars Kai Hansen, Professor DTU
Kim Steenstrup Pedersen, Professor, KU
Thomas Nikolajsen, Head of Front-end Innovation, FOSS
Toke Lund-Hansen, Head of Spectroscopy Team, FOSS
Erik Schou Dreier, Senior Scientist, FOSS

What is a Deep Learning Neural Network?

Deep learning neural networks are computer systems inspired by how our brains function. It consists of artificial neurons called nodes organized in layers. Each node takes in information, processes it, and passes it on to the next layer. This helps the network understand data and make predictions. By training the network with examples and adjusting the connections between nodes, it learns to make accurate predictions on new data. Deep learning neural networks are used for tasks such as image recognition, language understanding, and problem-solving.

Tags Impact story, p01

Future of work News

Cyber systems collaboration could help bring “lab on a chip” into the real world

Post author By Susanne Brøndberg
Post date June 13, 2023

13 June 2023

Cyber systems collaboration could help bring "lab on a chip" into the real world

By Søren Bjørn-Hansen

Getting a complex piece of hardware to work with all the variables of the real world is difficult. But through a DIREC-project collaboration, Luca Pezzarossa got much closer to having a working prototype.

Photo: Bax Lindhardt

For most people the term digital microfluidics doesn’t mean a whole lot. But it’s a technology which could revolutionise lab work completely.

“The idea is that whatever a biochemist normally spends a lot of time doing with pipettes, can be done by a chip instead,” Luca Pezzarossa explains and plays a short video of the technology at work.

The lab on a chip is basically a biochemical lab scaled down to the size of a small portable device. The idea is to make tiny droplets of fluid move around on a chip by activating a sequence of electrodes. Luca Pezzarossa’s job is to make the hardware and software work together to move the microfluidic droplets the right way.

“But one thing that is very common in this field is that the people who do this research are not biochemists, ” he says.

The problem with this is that there are real-world constraints, which makes moving the droplets much harder.

“Two droplets cannot go as close as they do in a simulation, because they would touch and merge. Or from a biological point of view, some droplets might leave behind a contaminating residue. Blood that leaves a trail for example, and so other droplets cannot move where this has moved earlier,” Luca Pezzarossa explains.

This represents a difficult problem. To translate from a high-level protocol that is user-friendly – something which is useful for a biochemist – down to a controlling sequence is very difficult. Especially when the constraints are complex real-world issues, like a droplet leaving a contaminating trail of blood on the chip.

“When I presented these challenges at a DIREC seminar, two algorithm-oriented scientists said: ‘This is very cool from a theoretical point of view. We should apply for a DIREC Explore project.’ And we did,” Luca Pezzarossa says.

Digital Research Centre Denmark (DIREC) supports multidisciplinary research – often with external partners – with the so called Explore projects, which are small agile research projects with the purpose of screening new ideas.

So for most of 2022 he worked with two fellow assistant professors, Eva Rotenberg from DTU Compute and Lene Monrad Favrholdt from University of Southern Denmark, on developing algorithms that could do this scheduling and routing on applied cyber-physical systems – while DTU research assistant Kasper Skov Johansen, who is now a PhD student at DTU, did most of the practical work.

Photo: Bax Lindhardt

“We ended up building a mathematical framework to describe the constraints. What are the different boundaries and rules these droplets need to respect?”, he explains.

They have laid out the ideas and know where they want to take the project next. Now they are looking for funding. Long term their goal is to make the lab on a chip better so it works well in the real world instead of in a simulation.

Luca Pezzarossa believes a prototype is a couple of years away. But it all depends on how the users of the lab on a chip react.

“When people start using something they will tell you ‘this shouldn’t behave like that’. But that is the point. To try to do something useful and not just move coloured droplets around. Some issues you only discover when you bring things into the real world,” he says.

Through their collaboration, the team discovered that developing the right algorithms was much more complex than they thought. They only partially succeeded in creating the right algorithms. And that is okay, he thinks:

“It’s a difficult problem to solve. And that is also the purpose of DIREC Explore. You explore an idea, and you see if there is more that you can do. And in this case, there was more. Which is why we are looking for more funding.”

According to Luca Pezzarossa the DIREC Explore project was essential because it got them thinking differently.

“We tried to bring two worlds together that are very different. I’m from the embedded systems world, which is very noisy and real. We build things inside rockets and cars. But the world of algorithms is very formal and abstract at the same time. And this, I think, was one of the aims of the explore project, to bring people together from different fields,” he says and continues:

“I learned a lot about communication and the need to explain things in a common language. It is sometimes difficult, but so worth it. Otherwise, you are in your perfect box with all the things you know. Which is also good research, in principle. But the real world isn’t really made out of clean boxes.”

Luca Pezzarossa is an assistant professor at DTU Compute. He first arrived at DTU from his native Italy on an Erasmus Exchange Program in 2012. He did his master thesis at DTU two years later and has been at DTU ever since.

Tags Impact story, p27

Future of work Green Tech News

Data detects irregularities before things go wrong

Post author By Susanne Brøndberg
Post date May 25, 2023

25 May 2023

Data detects irregularities before things go wrong

A defect at a processing plant in Brazil meant that production was at a standstill for three days. The incident has prompted SANOVO TECHNOLOGY GROUP to invest time and data in a DIREC research project, which involves machine learning and IoT with the aim of preventing similar breakdowns in the future.

Every minute was crucial when a critical machine component failed, requiring the new replacement part to be shipped from SANOVO in Denmark to Brazil. During this time, the sorting plant was at a standstill.

At SANOVO TECHNOLOGY GROUP, one of the world’s leading companies in the development and production of advanced machines and equipment for the egg industry, efforts are being made to avoid similar situations in the future.

Therefore, the company is participating in a project at the national centre for digital technologies (DIREC), where they, together with researchers from the University of Southern Denmark, Aalborg University, and the University of Copenhagen, are investigating how data can be used to detect even small deviations in a production facility.

– If we can somehow get a warning, for example, a month before something happens to a specific component, we can intervene faster and save the customer from the production line coming to a halt,” says Steven Beck Klingberg, System Manager at SANOVO TECHNOLOGY GROUP.

We can probably save a lot of money on travel activities, but otherwise, it will have a significant impact on our customers. If a machine is idle for a week, it can cost the customer several hundred thousand euros in lost production.
– Steven Beck Klingberg, System Manager at SANOVO TECHNOLOGY GROUP

Data reveals irregularities

The company extracts several hundred data points from systems around the world. So far, focus has been on production data, but recently, researchers have shifted their attention to data that reveals the machine’s condition, explains Professor Fabrizio Montesi from SDU, who leads the project.

“We use IoT, edge, and cloud technologies to accumulate data on the function of machines implemented in production and test environments. By analyzing this data, we identify conditions and trends that indicate deviation from normal function. This insight can then be used to predict when a machine needs servicing.”

His colleague on the project, Associate Professor Marco Chiarandini, adds:

– The uniqueness of SANOVO is that the amount of data is large, while errors in the main component are extremely rare. Therefore, classical monitoring and traditional machine learning techniques are not suitable, and we have had to tailor other data science techniques for sequential data analysis.

Aiming to reduce the frequency of maintenance travels

As a side benefit, the project may help reduce the number of maintenance travels, a goal that is important for SANOVO for both environmental and economic reasons.

The company has service personnel employed in Denmark, Holland, Italy, South and North America, Malaysia, Japan, and China – each department has its own area of expertise. A service technician has between 150 and 200 travel days per year, with the entire service organization totaling just over 100 employees.

– If we can predict that a machine will soon need servicing, it will be easier to plan service trips and minimize travel activity – and it will make a difference. We will not only have a better understanding of what is wrong before sending a service technician out into the world, so he can have the right machine parts with him. We also want to catch problems early on, so we can plan smarter and minimize the number of travels, says Steven Beck Klingberg.

Researchers and students dare to challenge

SANOVO’s role in the DIREC project is to contribute expertise on relevant machine data. There are several hundred measurement points in the machines, but not all are significant for the critical components of the machine.

– We have primarily helped researchers figure out which measurement points are important. In that way, we are sparring partners throughout the process, says Steven Beck Klingberg.

There is no doubt that the project is important for the company. Several of SANOVO’s specialists have been involved in the project, which is also followed with great interest by top management.

The collaboration between researchers and a highly specialized company brings a lot of new knowledge and ideas to the table, according to Steven Beck Klingberg.

– Both the researchers and the students we collaborate with are excellent at asking questions that challenge us, and it has been great to get other perspectives along the way. Researchers come with an open mindset and completely new knowledge. It has been fantastic to get some counterplay because you can become a bit narrow-minded when working with the same things in the same industry day in and day out.

The researchers also see great value in the collaboration.

– Identifying a project of concrete value to Sanovo has been the key to gaining support and interest from the right people in the company, which has been crucial for the success of the collaboration. All parties have been quite open in the research phase, and we all benefit from the new experience and knowledge exchange, says Fabrizio Montesi.

FACTS

SANOVO TECHNOLOGY GROUP is a world leader in process solutions for the egg industry but is also specialized in various other business areas such as enzymes, pharma, hatcheries, and spray drying of other protein sources.

The innovative engineering work for the egg industry began in 1961, and today, SANOVO TECHNOLOGY GROUP is a company with almost 600 employees and customers worldwide. With its own service and sales offices on six continents and production in Denmark, Holland, Slovakia, and Italy, SANOVO TECHNOLOGY GROUP is a global partner in the egg industry.

The overall purpose of the DIREC project ‘DeCoRe: Tools and Methods for the Design and Coordination of Reactive Hybrid Systems’ is to explore the applicability of technologies and methods for designing hybrid systems, including IoT, edge, and cloud solutions, in the industry.

The future hybrid meetings: Now you can "whisper" to your virtual colleague

A walkie-talkie for hybrid meetings

A collaboration between companies and researchers

Error source or necessity? Researchers explore the human role in future production

Collaboration between humanities and engineering

Humans play an essential role

Swarms of robots are being deployed on the fields – What does it take to expand the use of them?

The collaboration uncovers various aspects of the technology

ABOUT THE HERD PROJECT

For four decades, Susanne has been shaping the digital everyday life of Danes

Examine your organisation critically and inquisitively

DIREC Interview: Professor Henrik Christensen on developments in robot technology

Exhibition gave a twisted glimpse into the future workplace

Researchers and artists are designing the future hybrid workplace

Seeing the world differently

About the REWORK project

Explainable AI will disrupt the grain industry and give farmers confidence

Mapping and explaining algorithms

An important competitive advantage

Facts about FOSS

Deep Learning and Automation of Imaging-based Quality of Seeds and Grains

Cyber systems collaboration could help bring "lab on a chip" into the real world

Data detects irregularities before things go wrong

Data reveals irregularities

Aiming to reduce the frequency of maintenance travels

Researchers and students dare to challenge

FACTS

Contact info

Newsletter signup