Smart algorithms permeate deeper and deeper into our lives. They influence what music we listen to, where we eat and what we eat, what people we meet, whether we are invited for a job interview, what information we find, and so on. Soon advanced bots will 24/7 diagnose our health and advise us on nutrition and lifestyle. When will the personal care robot become available, that - when we get old and needy- helps us out of bed and under the shower, reminds us of regular eating, drinking and taking medication, supports our housekeeping and guides us during walks?
For people, emotions are essential in communication. If we find something important, we put an emotional load on it, and if we don't, it's apparently not important. We also see reasonably well how someone feels about what we do. Without emotions, we cannot set priorities, we can’t make good choices.
If we want people and (ro)bots to really work together, it is reasonable to assume that emotions will play an important role. Technology is getting ready for this. Computers are able to recognise and express emotions.
How could we make emotions an integral part of the interaction between people and bots, and how could this lead to better coordination and more trust? Do we need to mimic human emotions? How can we conduct research into this?

The main competitive advantage of brands like BMW and Tesla is not in the electrification of cars, but in the development of AI for autonomous driving based on massive amounts of data gathered by the cars during their use. IBM, Apple, and Google are miles ahead in areas like cognitive processing, search and language recognition because of their intensive deep learning approaches leveraging huge amounts of training data. And the list goes on. AI is the key technology and skill in many important and relevant application domains.

Compared to the USA and China, Europe is seriously lagging behind in the development of Artificial Intelligence.

However, there are hopeful developments in The Netherlands and more specific the Brabant region. Examples are the TU/e initiated the Eindhoven Artificial Intelligence Systems Institute (EAISI), the Jheronimus Academy of Data Science in Den Bosch, and Tilburg’s MindLabs. Dutch Government aims to direct 1+1 billion euro's to AI inspired innovation in the next 10 years and the recently launched Dutch AI Coalition (NLAIC) has been assigned with the challenging task to orchestrate the activities at national level that should lead to the desired socio-economic benefits.

Millions of kilometers have been driven with autonomous test vehicles and AutoPilot-like driver assistance functions are appreciated by many users every day. All big car manufactures invest heavily on automated driving and a world without severe accidents seems realistic within years. At the same the Dutch safety board recently concluded that the introduction of advanced driver assistance systems introduces serious risks on road safety. This conclusions seems to contradict promises we have heard for many years. What can we expect from our automated vehicles and what makes this problem complicated from a technological perspective?

This talk looks for an answer to this question by elaborating on one of the fundamental technologies required for vehicle automation: world modeling. A world model represents an automated vehicle’s surroundings and contains the vehicle’s own position, a road model and positions and intentions of other traffic participants. By explaining both required and available technology and algorithms we explore the current state of the art and the promising future of automated driving.

For the past 35 years, thousands of Dutch SMEs have kept their books in Exact software. To help them make the most out of their data, two years ago Exact has taken up the challenge of revolutionizing its software, introducing Machine Learning to support full automation of data entry and Data Science to deliver key financial insights and advise to each and every one of its 350000 customers.

This talk reviews how AI and machine learning can be applied in business software to empower businesses to make truly data-driven decisions, also going through real world examples of how Exact delivers rich financial insights and automation at scale for hundreds of thousands of Small and Medium Enterprises.

Do you want your loved ones to live forever? Well, maybe soon this will be possible. Sort of.

More and more we sent apps, use social media, listen to music, watch series and leave an impressive data trail. What if we could use that data to create a chatbot? A chatbot that lives on after someone had died. A griefbot.

Some prototypes are already on the market.

Do you think this is creepy? Or do you think it is a good idea? Should it be banned? Or stimulated? What is the impact on society? How should it be designed? For who?

Fontys is researching a Technology Impacting Cycle Tool to help you answer these questions and many more. The tool provides you with an awareness framework that helps you to ask and answer the right questions.

In this talk we will use the example of the Griefbot to show you how using this tool helps you to improve your thinking about the impact of technology.

The tool is already operational in beta, so after the talk the audience can start working with the tool themselves.

During this talk we will elaborate on the work we did at LABS, the explorative innovation hub of Greenhouse Group. We explored generative models and their possible applications. Creativity is typically not thought of as a quality or skill a computer could learn. However we see that recent advances in AI now make it possible to generate interesting content. The goal of our research is to find ways on how human and machine could work together in the future to perform creative tasks.

At Prodrive Technologies we believe in automation, automation of everything. Not only for our customers, but also for our own manufacturing facilities. The Prodrive Robotics Solutions (PRS) group develops generic robotic automation cells in-house that can work autonomously and robustly on many different products requiring no manual intervention.

Vision camera systems and algorithms, using 2D and 3D data, are developed for (1) detection and localization of components for automated pick and place actions, and (2) automated visual inspection. Because high-end electronics are manufactured, submillimeter accuracy is required for placement actions. Components that must be detected and localized range from small lenses with a width of just a few millimeters, up to housings of several hundred millimeters.

Automated visual inspections are performed to validate manufacturing processes and to ensure the highest production quality. Inspection steps range from sealant inspection, where dispensed material such as a glue is verified on position and thickness, to presence detection of components and correctness of wire placement. For this last inspection, deep learning algorithms are used to cope with the large variety of input data.

To speed up the integration and commissioning time, design files (CAD) can be used as input of machine vision algorithms as well. These files are used to generate data sets in a simulated environment for training and validation of machine vision and deep learning algorithms without requiring any manufacturing equipment or components.

What sets robots apart from other intelligent agents, is that they’re capable of physical movement. As people are highly sensitive to physical movement and spatiotemporal affordances, we argue that robot motion can go a long way towards people’s comprehension and acceptance of robots. In the field of animation a body of knowledge has been established on how to convey any idea, message or intention via motion. The idea of incorporating animation is therefore not new to robotics. However, it is barely visible in today’s world, because the adoption of animation has only been investigated in social robotics. As a result, the growing numbers of implemented robots outside of that category are left to rely on traditional signals like light and sound for communicating.

I will discuss three concepts from animation that can be adopted to create intuitively interpretable robot behaviour. To illustrate these and the importance of a more movement-centric design approach, a case study of a hospital transport robot will be presented, showing that this approach cannot only contribute to the acceptance of a robot and a more pleasant environment, but also to a more efficient system.

At Prodrive Technologies we believe in automation, automation of everything. Not only for our customers, but also for our own manufacturing facilities. The Automated Guided Vehicle (AGV) is an important project in realizing our automation dream; it is the missing link between all other automated processes.

The AGV is a complex system which is in-house developed, manufactured, and also used. Developing an AGV for environment where humans work gives unique challenges. Deploying an AGV in such environment results in interesting scenario’s and shows how humans and robots can collaborate successfully.

The AGV is a versatile vehicle which is able to automatically transport payloads from a given origin to a destination. The AGV is able to do this efficiently in diverse and changing environments. Using multiple AGVs together in one environment imposes extra challenges, which are solved by our Fleet Management System (FMS). Controlling and retrieving statistics from an AGV is done with use of the dashboard, which is the graphical user interface of the AGV solution.

Smart Industry / Industrie 4.0 is about the digitization of companies. This impacts the boundaries between the physical, digital, logistical and human worlds. It changes the way we produce by integrating sensors, industrial IoT, with the way we process and integrate data. Trust is key for the smart industry. For example, when driving with automated guided vehicles, when assembling complex products with robot arms and when navigating intelligent parts along the assembly line.
This presentation is about the question how to do smart industry from a practical research oriented view. It does so by taking examples at hand from several project conducted by the research group High Tech Embedded Software of Fontys ICT. Attention will be paid to driving with autonomous vehicles on the shop floor, human robot interaction, and cybersecurity.

Genuine Empathic technology is getting under your skin. The success of games like “Keep talking, nobody explodes” or "The Werewolves of Miller's Hollow“ lies in the excitement of the emotional response from the players. However, the game does not adapt to it. Current technology promises that it can. Several software packages are quite advance in recognizing a single indicator of an emotional response, for example facial expression, voice-pitch and/or heartbeat. In our latest research we combined, integrate and applied these technologies in a game context.

FHICT-Game Design explored the opportunities from three different angles: Information display, in entertainment games and in an applied game for first responders.

1. Displaying emotional information in real-time has it’s challenges when applied in a live context. It should be presented unobtrusive and it should be clear enough for immediate interpretation. How do you design an emotional feedback loop?

2. Many games elicit excitement, some demand to hide it (such as poker), some thrive by the expression of it as in party games. There are not so many games that demand to control and manipulate emotions per sé. What if we could?

3. First responders need to rely on each-other’s emotional stability in crisis-situations. This is an aspect that is difficult to train with current training materials. It was technically difficult to train and test emotional stability in real-time. Therefore we designed a real-time Emotional Game Response in a VR-Narrative.
In this presentation we will present each case and share the lessons learned.

The way robot colleagues, aka co-bots, function/operate on the workfloor is not always acceptable yet for their human colleagues. They behave more like machines than team members. Trust is a very important aspect in teamwork. Research shows that trust between people develops if we can rely on the other person, if his/her behavior is logical and predictable and if he/she means the best to us.

In this talk we will show how one can improve the communication of the collaborative industrial robot to make it more legible (understandable) and predictable. We will demonstrate that this may be done with subtle, minimalistic means and adjustments, e.g. in its appearance and non-verbal behavior. This way our robot colleagues can become more human friendly.

A Case Study of Communication Skills Training in E-Learning.

Emotion recognition technology is a high tech approach that allows computers to read the emotional states of humans, and provide them with the appropriate feedback. Moreover, we recognise this technology as artificial emotional intelligence (AEI), or Emotion AI.
Humans communicate with each other through nonverbal communication approach to send and receive messages without using any words, either written or spoken. Different nonverbal methods are called facial expressions, vocal intonations, body gesture, body language, etc.
In this talk, we focus on emotion recognition from facial expressions using webcam data to provide real-time, continuous, and unobtrusive recognition of facial expressions.
Furthermore, we will show you how to use this technology in an e-learning environment to improve communication skills training sessions for university students and professors.

In April 2018, the month Avicii died, I tried to get a new understanding of the developments in his music. The journey started as a research project in which I leverage the Google Cloud Natural Language API to analyse songs for sentiment. It ended as a unique way to visualise emotion in music.

From the currently running various cyber security R&D projects it shows that steps are taken towards a better and more mature security in smart industry. However, a lot of improvements are still needed. We present the status and results of our latest cyber security R&D projects related to smart industry.

HERE Technologies uses 6 Cobots to test their navigation software on real Navigation systems as used by their automotive OEM customers.

Each Cobot has received a “human” name, like: Dennis, Angela or Eddie, and performs a task which could also be done by a human.

The cobots are becoming part of the department organigram and available resources for planning tools.

Fontys Applied Data Science minor project team 4 has started the investigation how to make the cobots more smart by using AI and ML technologies. The 1st outcome is that using ML can reduce the amount for false positives and false negatives.

It is evident that the current cobots reject positive results, including good ones, based on a very simple if then equation whereas a human would accept the result.

The very difference is that humans can react instantly on very small and sudden changes, whereas a cobot can only reply in OK or Not OK.

The minor project proved that intelligence can be achieved by implementing Reinforcement Learning and recommended to use a 3 stage action program in the robot:

1st : Problem occurrence identification
2nd: Problem definition
3rd: solving the problem.

A lot of (big) data is still based on citizens, clients or patients filling in Questionnaires. The famous customer or employee satisfaction surveys, PREM’s, PROM’s and other outcome surveys are the basis for policy, productdesign, research into new medicines and the effectiveness of treatment. But often you can’t capture human feelings or experiences in numbers and of course with all surveys, they are prone to socially desirable answers and non-response.

We are investigating new possibilities of gaining insights into what people really think, feel and dream. Chatbot’s might be a good way to start!

Is AI really going to cause an industrial revolution scale change, or is it just the next hyped tech buzzword like blockchain? In this talk we will briefly go over AI and break it up into technical components that actually have a clear definition. From this point we will look at how AI is presented in the news and blog posts and evaluate where it is actually implemented in everyday industry and business. You will quickly find out that we are far from using AI as effectively as the news might portray, but also that the underlying technology has promising applications in some areas

Human ethics and machine-ethics are two different but related things. I will adress the issue of moral programming and how it relates to ancient moral philosophy. Deontic logic is a tool that can be used to program the more absolutist moral approaches, which is pretty much the ethics people do not do. Therefore we need to rethink our Ethical Desicion Making Models (EDMs) when we try to answer the following questions:
1) who decides on moral programming of smart technology
2) how do we want it to be programmed when it interacts with us

TamaCoachee is a play on words. It’s a reference to the Tamagotchi of the ‘90s where you had to take care of a virtual animal. With the TamaCoachee you take care of a virtual coachee by being it’s/his/her coach. It’s meant as a training tool for students of Applied Psychology (and related studies) to practice coaching skills. It will look similar to using Skype for a video call, but with feedback on your interaction. We will present an early interactive prototype of this amazing new tool.

Virtual humans (Siri, Alexa, etc.) are ‘virtual humans’ that assist us in a diverse range of tasks. Virtual humans can and are being used to train social skills. Managers, psychologists, coaches, negotiators, etc. can practice on with virtual humans during their training. The benefits are great: detailed insight in participants progress, lower costs and a more homogenous experience per student compared to role playing. One of the main points of focus of the TamaCoachee is the psychological impact of ‘connecting’ real and virtual humans.

Many people and companies perceive AI as the new electricity – a powerful and mature technology that is almost a given. And this view is not surprising, because algorithms are already transforming our lives. They help us to get from A to B, recommend us things to buy or watch, find fraudulent transactions, and predict how we vote, or even with whom we will fall in love. But can it really be commercialized ‘just like that’ to create smart innovative solutions for society?

My answer to this question is a firm and clear-cut “no”. In this talk I want to separate the hype from reality, and sketch what still needs to be done in the field of AI.

Engineering machine learning applications is different from engineering traditional rule-based software. We do not specify the rules ourselves, but instead deploy self-learning algorithms that discover the rules from data. So we need mechanism to explicit these learned rules. This is useful for end users affected by the algorithms but also for engineers that need to debug the algorithms. This presentation focusses on the latter. I will explain the challenges that engineers face when engineering machine learning applications and present tools and solutions available to face those challenges.

To address challenges and demands from current society, life long learning is necessary for each citizen to keep up to the changing demands on knowledge and skills. More and more, skills development and knowledge acquisition takes place outside a formal educational institution (e.g. on the workplace or in environments like sports clubs or theater). For educational institutions this demands a high rate of flexibility in programs. One of the measures to realize this is to recognize and reward smaller parts of a program (micro-credentialing). Fontys School of ICT therefore is developing a vision on this phenomenon. This vision addresses questions like: what requirements are minimal needed for recognition of skills and knowledge or how to recognize skills gathered in a non-formal setting. In this presentation we will present some backgrounds on micro-credentialing and the current state of the vision. We will then discuss with participants about their ideas of micro-credentialing and open badges.