European Robotics Forum 2017

The European Robotics Forum (ERF2017) took place between 22 and 24 March 2017 at the Edinburgh International Conference Centre.

The goals were to:

  • Ensure there is economic and societal benefit from robots
  • Share information on recent advancements in robotics
  • Reveal new business oppourtunities
  • Influence decision makers
  • Promote collaboration within the robotics community

The sessions were organised into workshops, encouraging participants from academia, industry and government to cross boundaries. In fact, many of the sessions had an urgent kind of energy, with the focus on discussions and brainstorming with the audience.

wp-1490897576308.jpg

Edinburgh castle at night

Broad spectrum of robotics topics

Topics covered in the conference included: AI, Social Robotics, Space Robotics, Logistics, Standards used in robotics, Health, Innovation, Miniturisation, Maintenance and Inspections, Ethics and Legal considerations. There was also an exhibition space downstairs where you could mingle with different kinds of robots and their vendors.

The kickoff session on the first day had some impressive speakers – leaders in the fields of AI and robotics, covering business and technological aspects.

Bernd Liepert, the head of EU Robotics covered economic aspect of robotics, stating that the robot density in Europe is around the highest in the world. Europe has 38% of the world wide share of the professional robotics domain, with more startups and companies than the US. Service robotics already makes over half the turnover of industrial robotics. In Europe, since we don’t have enough institutions to develop innovations in all areas of robotics, combining research and transferring to industry is key.

The next speaker was Keith Brown, the Scottish secretary for Jobs, the Economy and Fair Work, who highlighted the importance of digital skills to Scotland. He emphasised the need for everyone to benefit from the growth of the digital economy, and the increase in productivity that it should deliver.

Juha Heikkila from the European Commission explained that, in terms of investment,  the EU Robotics program is the biggest in the world. Academia and industry should be brought together, to drive innovation through innovation hubs which will bring technological advances to companies of all sizes.

wp-1490898806984.jpg

Raia Hadsell of Deep Mind gave us insight into how deep learning can be applied to robotics. She conceptualised the application of AI to problem areas like speech and image recognition, where inputs (audio files, images) are mapped to outputs (text, labels). The same model can be applied to robotics, where the input is sensor data and the output is an action. For more insight, see this article about a similar talk she did at the Re•Work Deep Learning Summit in London. She showed us that learning time can be reduced for robots by training neural networks in simulation and then adding neural network layers to transfer learning to other tasks.

Deep learning tends to be seen as a black box in terms of traceability and therefore risk management, as people think that neural networks produce novel and unpredictable output. Hadsell assured us, however, that introspection can be done to test and verify each layer in a neural network, since a single input always produces a range of known output.

The last talk in the kickoff, delivered by Stan Boland from Five AI, brought together the business and technical aspects of self driving cars. He mentioned that the appetite for risky tech investment seems to be increasing, with a 5 times growth in investment in the past 5 years. He emphasised the need for exciting tech companies to retain European talent and advance innovation, and reverse the trend of top EU talent migrating to the US.

On the technology side, Stan gave some insight into some advances in perception and planning in self driving cars. In the picture below, you can see how stereo depth mapping is done at Five AI, using input from two cameras and mapping the depth of each pixel in the image. They create an aerial projection of what the car sees right in front of it and use this birds eye view to plan the path of the car from ‘above’. Some challenges remain, however, with 24% of cyclists still being misclassified by computer vision systems.

With that, he reminded us that full autonomy in self driving cars is probably out of reach for now. Assisted driving on highways and other easy-to-classify areas is probably the most achievable goal. To surpass this, the cost to the consumer becomes prohibitive, and true autonomous cars will probably only be sustainable in a services model, where the costs are shared. In this model, training data could probably not be shared between localities, with very specific road layouts and driving styles in different parts of the world (e.g Delhi vs San Francisco vs London).

This slideshow requires JavaScript.

An industry of contrasts

This conference was about overcoming fragmentation and benefitting from cross-domain advances in robotics, to keep the EU competitive. There were contradictions and contrasts in the community which gave the event some colour.

Each application of robotics that was represented seemed to have its own approaches, challenges, and phase of development, like drones, self driving cars, service robotics and industrial robotics. In this space, industrial giants find themselves collaborating with small enterprises – it takes many different kinds of expertise to make a robot. The small companies cannot afford to spend the effort that is needed to conform to the industry standards while the larger companies would go out of business if they did not conform.

A tension existed between the hardware and software sides of robotics – those from an AI background have some misunderstandings to correct, like how traceable and predictable neural networks are. The ‘software’ people had a completely different approach to the ‘hardware’ people as development methodologies differ. Sparks flew as top-down legislation conflicted with bottom-up industry approaches, like the Robotic Governance movement.

The academics in robotics sometimes dared to bring more idealistic ideas to the table that would benefit the greater good, but which might not be sustainable. The ideas of those from industry tended to be mindful of cost, intellectual property and business value.

Two generations of roboticist were represented – those who had carried the torch in less dramatic years, and the upcoming generation who surged forward impatiently. There was conflict and drama at ERF2017, but also loads of passion and commitment to bring robotics safely and successfully into our society. Stay tuned for the next post in which I will provide some details on the sessions, including more on ethics, legislation and standards in robotics!

Making social robots work

 

wp-1490820852441.jpg

Mady Delvaux, in her draft report on robotics, advises the EU that robots should be carefully tested in real life scenarios, beyond the lab. In this and future articles, I will examine different aspects of social robot requirements, quality and testing, and try to determine what is still needed in these areas.

Why test social robots?

In brief, I will define robot quality as: does the robot do what it’s supposed to do, and not do what it shouldn’t. For example, when you press the robot’s power button from an offline state, does the robot turn on and the indicator light turn green? If you press the button quickly twice, does the robot still exhibit acceptable behaviour? Testing is the activity of analysis to determine the quality level of what you have produced – is it good enough for the intended purpose?

Since social robots will interact closely with people, strict standards will have to be complied with to ensure that they don’t have unintended negative effects. There are already some standards being developed, like ISO13482:2014 about safety in service robots, but we will need many more to help companies ensure they have done their duty to protect consumers and society. Testing will give insight into whether these robots meet the standards, and new test methods will have to be defined.

What are the core features of the robot?

The first aspect of quality we should measure is if the robot fulfils its basic functional requirements or purpose. For example, a chef robot like the robotic kitchen by Moley would need to be able to take orders, check ingredient availability, order or request ingredients, plan cooking activities, operate the stove or oven, put food into pots and pans, stir, time cooking, check readiness, serve dishes and possibly clean up.

 

A robot at an airport which helps people find their gate and facilities must be able to identify when someone needs help, determine where they are trying to go (perhaps by talking to them, or scanning a boarding pass), plan a route, communicate the route by talking, indicating with gestures, or printing a map, and know when the interaction has ended.

 

With KLM’s Spencer the guide robot at Schiphol airport, benchmarking was used to ensure the quality of each function separately. Later the robot was put into live situations at Schiphol and tracked to see if it was planning movement correctly. A metric of distance travelled autonomously vs non autonomously was used to evaluate the robot. Autonomy will probably be an important characteristic to test and to make users aware of in the future.

Two user evaluation studies were done with Spencer, and feedback was collected about the robot’s effectiveness at guiding people around the airport. Some people, for example, found the speed of the robot too slow, especially in quiet periods, while others found the robot too fast, especially for families to follow.

Different environments and social partners

How can we ensure robots function correctly in the wide variety of environments and interaction situations that we encounter everyday? Amazon’s Alexa, for example, suffers from a few communication limitations, like knowing if she is taking orders from the right user and conversing with children.

At our family gatherings, our Softbank Nao robot, Peppy, cannot quite make out instructions against talking and cooking noises. He also has a lot of trouble determining who to focus on when interacting in a group. Softbank tests their robots by isolating them in a room and providing recorded input to determine if they have the right behaviour, but it can be difficult to simulate large public spaces. The Pepper robots seem to perform better under these conditions. In the Mummer project, tests are done in malls with Pepper to determine what social behaviours are needed for a robot to interact effectively in public spaces.

 

The Pepper robot at the London Science Museum History of Robots exhibition was hugely popular and constantly surrounded by a crowd – it seemed to do well under these conditions, while following a script, as did the Pepper at the European Robotics Forum 2017.

When society becomes the lab

Kristian Esser, founder of the Technolympics, olympic games for cyborgs, suggests that in these times, society itself becomes the test lab. For technologies which are made for close contact with people, but which can have a negative effect on us, the paradox is that we must be present to test it and the very act of testing it is risky.

Consider self-driving vehicles, which must eventually be tested on the road. The human driver must remain aware of what is happening and correct the car when needed, as we have seen in the case of Tesla’s first self driving car fatality: “The … collision … raised concerns about the safety of semi-autonomous systems, and the way in which Tesla had delivered the feature to customers.” Assisted driving will probably overall reduce the number of traffic-related fatalities in the future and that’s why its a goal worth pursuing.

For social robots, we will likely have to follow a similar approach, first trying to achieve a certain level of quality in the lab and then working with informed users to guide the robot, perhaps in a semi-autonomous mode. The perceived value of the robot should be in balance with the risks of testing it. With KLM’s Spencer robot, a combination of lab tests and real life tests are performed to build the robot up to a level of quality at which it can be exposed to people in a supervised way.

Training robots

Over lunch the other day, my boss suggested the idea of teaching social robots as we do children, by observing or reviewing behaviour and correcting afterwards. There is research supporting this idea, like this study on robots learning from humans by imitation and goal inference. One problem with letting the public train social robots, is that they might teach robots unethical or unpleasant behaviour, like in the case of the Microsoft chatbot.

To ensure that robots do not learn undesirable behaviours, perhaps we can have a ‘foster parent’ system – trained and approved robot trainers who build up experience over time and can be held accountable for the training outcome. To prevent the robot accidentally picking up bad behaviours, it could have distinct learning and executing phases.

The robot might have different ways of getting validation of its tasks, behaviours or conclusions. It would then depend on the judgement of the user to approve or correct behaviour. New rules could be sent to a cloud repository for further inspection and compared with similar learned rules from other robots, to find consensus. Perhaps new rules should only be applied if they have been learned and confirmed in multiple households, or examined by a technician.

To conclude, I think testing of social robots will be done in phases, as it is done with many other products. There is a limit to what we can achieve in a lab and there should always be some controlled testing in real life scenarios. We as consumers should be savvy as to the limitations of our robots and conscious of their learning process and our role in it.