Tuesday - May 30, 2017

Venue: Grand Ballroom (Level 5)

TIME	SESSIONS
0845 – 0945	PLENARY 1 Modeling the possibilities: From the Chalkboard to the Race Track to the World Beyond Chris Gerdes, Stanford University, USA
1345 – 1415	KEYNOTE 1 EndoMaster: A Surgical Robot’s Journey from the Research Lab to the Operating Theatre Louis Phee, Nanyang Technological University, Singapore
1415 – 1445	KEYNOTE 2 Capturing Vivid 3D Models of the World from Video Lourdes Agapito, University College London, UK

Wednesday - May 31, 2017

Venue: Grand Ballroom (Level 5)

TIME	SESSIONS
0820 – 0920	PLENARY 2 Nobel Turing Challenge: Grand Challenge of AI, Robotics, and Systems Biology​ Hiroaki Kitano, Sony Computer Science Laboratories, Inc., Japan
1230 – 1300	KEYNOTE 3 Industry 4.0 – Automation and Robotics Peter Luh, University of Connecticut, US
1300 – 1330	KEYNOTE 4 Research at the Intersection Between Robots and Play: Designing Robots for Children’s Healthcare Ayanna Howard, Georgia Tech, USA

Thursday - June 1, 2017

Venue: Grand Ballroom (Level 5)

TIME	SESSIONS
0820 – 0920	PLENARY 3 Framing the International Discussion on the Weaponization of Increasingly Autonomous Technologies Kerstin Vignard, United Nations Institute for Disarmament Research
1450 – 1520	KEYNOTE 5 An Operational Platform of Cloud Robotics Bill Huang, Cloud Minds, China
1520 – 1550	KEYNOTE 6 Model-based Optimization for Humanoid and Wearable Robots​ Katja Mombaur, University of Heidelberg, Germany

Chris Gerdes, Stanford University, USA

Modeling the possibilities: From the Chalkboard to the Race Track to the World Beyond

Simple mathematical models of physical systems give us tremendous insight into the nature of their underlying dynamics and the control challenges they present. Stripping away unnecessary details can illuminate fundamental dependencies and focus engineering efforts on the most critical problems. The challenge comes when models become so familiar that they are no longer taken as simplifications but mistaken for reality itself. Opportunities and possibilities that lie outside the bounds of those simple models are subsequently missed.

Ground vehicle dynamics represent an ideal illustration of these principles. While vehicles are complex multi-body systems with uncertain and nonlinear dynamic properties due to tire mechanics, many simplifications of these dynamics exist. By choosing the right level of abstraction for a model, anything from parallel parking to a race car driver’s choice of trajectory to drifting with smoking rear tires can be explained clearly and concisely. The choice of model is important, however, since what one model illuminates, another may obscure. This talk demonstrates through mathematics and video from experiments how simple models can be used to accurately control automated vehicles through even the most extreme maneuvers on the race track. Lap times comparable to expert drivers and drifting maneuvers beyond the precision of a human are possible with models consisting of only a few state variables. 

Just as models can guide or limit us in our work as researchers and engineers, our models of what it means to be a researcher or academic can sometimes artificially limit our impact in the world. The talk concludes with some simple models of how the robotics community can provide necessary leadership and technical guidance as society wrestles with the changes arising from our technologies. 

Hiroaki Kitano, Sony Computer Science Laboratories, Inc., Japan

Nobel Turing Challenge: Grand Challenge of AI, Robotics, and Systems Biology

Nobel Turing Challenge is one of the ultimate challenges the scientific community can tackle. It aims at (1) developing AI system including substantial robotics components that can make major scientific discoveries some of which worth Nobel Prize (called as “Scientific Discovery Challenge”), and (2) actually win the prize without the selection committee noticed that it is actually an AI system, not a human researcher (Cybernetic Personality Challenge). Primary focus on this challenge will be biomedical science area for Physiology and Medicine Award (Kitano, H., AI Magazine, 37(1) 2016).

This grand challenge project shall take a form of globally distributed “Virtual Big Science” project (Kitano, H., et al., Nature Chemical Biology, 7, 323-326, 2011). A part of the project shall resemble RoboCup (Kitano, H., et al., AI Magazine, 18(1) 73-85, 1997), but it will have substantially different aspects reflecting the difference of domains and objectives.

In the mid 90s, I advocated “Systems Biology” with the aim of promoting systems-oriented view in biology and to introduce more systematic measurements, proper applications of engineering, mathematical, and information science principles into life science (Kitano, H., Science, 295, 1662-1664, 2002; Kitano, H., Nature, 420, 206-210, 2002). This endeavor has been successful and systems biology is one of normal approach in biomedical and pharmaceutical sciences. The progress in systems biology revealed new limitations in life science that stems from our cognitive limitations to understand complex, non-linear, high dimensional, and dynamical systems, with overwhelming data and publications each of which unveils only a fragment of systems.

With recent breakthroughs in AI, exponentially increasing data production capabilities, and massive computing power, disruptive innovations in biomedical sciences are on the horizon. Time is ripe to embark on a new aggressive challenge. The fundamental breakthrough will come at the stage AI to generate hypotheses and quickly verify them using their knowledge bases, simulation, and robotics experimental systems. It means that AI systems can keep discovering new knowledge with minimal or zero human interventions. Even a mid-term achievement of Scientific Discovery Challenge alone will be a game changer. It will trigger fundamental transformations of industry and more largely on the shape of our civilization.

Kerstin Vignard, United Nations Institute for Disarmament Research

Framing the International Discussion on the Weaponization of Increasingly Autonomous Technologies

There are a multitude of positive military applications for increasingly autonomous technologies. However, their potential weaponization raises a host of legal, technical, operational and ethical questions. Since 2013, member states of the United Nations have been discussing the weaponization of increasingly autonomous technologies (Lethal Autonomous Robots, Lethal Autonomous Weapon Systems, or so-called “killer robots”) in both human rights and arms control fora. Four years in, there is still great division on definitions, how to ensure human control over these future weapon systems, and the appropriate policy responses.

These political discussions are held in the near absence of the technical community. As the rate of technological innovation far outpaces the policy discussion, how might engagement with the technical experts enable international policy-makers to better think, discuss and make informed decisions about increasing autonomy in weapon systems?

Louis Phee, Nanyang Technological University, Singapore

EndoMaster: A Surgical Robot’s Journey from the Research Lab to the Operating Theatre​

I will share my experiences in developing a novel flexible robotic system that removes gastric and colon tumours using natural orifices as points of access. I will discuss the technical and medical challenges faced to push the research to successfully test the robot on human subjects. Thereafter, a company (EndoMaster) was incorporated to commercialise the product. The challenges faced in the translation of the robotic technology to be used in a clinical setting were entirely different from the research phase. By sharing my experiences, I hope to inspire more researchers to translate their research and inventions to actual useful products.

Bio: Dr Louis Phee is a Professor at Nanyang Technological University (NTU), Singapore. He is Chair of the School of Mechanical & Aerospace Engineering at NTU. He graduated from NTU with the B.Eng (Hons) and M.Eng degrees in 1996 and 1999 respectively. He obtained his PhD from Scuola Superiore Sant’Anna, Pisa, Italy in 2002 on a European Union scholarship. His research interests include Medical Robotics and Mechatronics in Medicine. He was the founding CEO of EndoMaster Pte Ltd, a company he co-founded to commercialize a surgical robotic system he developed.

Dr Phee was awarded the Young Scientist Award (2006), the Outstanding Young Persons of Singapore Award (2007), the Nanyang Outstanding Young Alumni Award (2011), Nanyang Innovation and Entrepreneurship Award (2013) and the President’s Technology Award (2012). In 2005, he was awarded the Best Paper Award at the prestigious IEEE International Conference on Robotics and Automation.

Lourdes Agapito, University College London, UK

Capturing Vivid 3D Models of the World from Video

As humans we take the ability to perceive the dynamic world around us in three dimensions for granted. From an early age we can grasp an object by adapting our fingers to its 3D shape; we can understand our mother’s feelings by interpreting her facial expressions; or we can effortlessly navigate through a busy street. All of these tasks require some internal 3D representation of shape, deformations and motion.

Building algorithms that can emulate this level of human 3D perception has proved to be a much harder task. In this session, I will show progress from early systems which captured sparse 3D models with primitive representations of deformation towards the most recent algorithms which can capture every fold and detail of hands or faces in 3D using as input video sequences taken with a single consumer camera. There is now great short-term potential for commercial uptake of this technology, and I will show applications to robotics, augmented and virtual reality and minimally invasive surgery.

Peter Luh, University of Connecticut, US

Industry 4.0 – Automation and Robotics​

Industry 4.0 is a confluence of trends and technologies pushed by the digital revolution and the “Internet of Things,” and driven by customer demand for high quality and customized products at reasonable prices. With (1) the ubiquitous connection and interaction of machines, things and people, (2) the mixing of physical and virtual worlds, and (3) the emerging of disruptive technologies such as 3D printing and robotics, the ways we design and manufacture products and provide services will be fundamentally changed. In this talk, Industry 4.0 will be introduced, including its design principles and key technologies.  An important but difficult technology often missing in discussions - mathematical optimization, especially for problems involving discrete decision variables or for methods to coordinate distributed and autonomous “things,” will be highlighted. Applications to “Clean Energy Smart Manufacturing” will be presented, and implications on automation and robotics will be discussed. Finally, Industry 4.0 in the US, Europe, China and Japan will also be introduced. 

Ayanna Howard, Georgia Tech, USA

Research at the Intersection Between Robots and Play: Designing Robots for Children’s Healthcare

There are an estimated 150 million children worldwide living with a disability. For many of these children, physical therapy is provided as an intervention mechanism to support the child’s academic, developmental, and functional goals from birth and beyond. With the recent advances in robotics, therapeutic intervention protocols using robots is now ideally positioned to make an impact in this domain. There are numerous challenges though that still must be addressed to enable successful interaction between patients, clinicians, and robots - developing interfaces for clinicians to communicate with their robot counterparts; developing learning methods to endow robots with the ability to playfully interact with the child; and ensuring that the robot can provide feedback to the parent and clinician in a trustworthy manner.

I will discuss the role of robotics and related technologies for pediatric therapy and highlight our methods that bring us closer to this goal. I will present our approaches and preclinical studies in which these technologies address real-life developmental goals for children with special needs.

Bill Huang, Cloud Minds, China

An Operational Platform of Cloud Robotics

Cloud Robotics is a kind of robot whose “brains” are on cloud, with a global high-speed backbone network as the “nerve”. In the near future, all the intelligent robots will be Cloud Robotics. At this session, we will introduce the architecture of Cloud Robotics, and how to build an operational platform for millions of robots with human assistant robot intelligence, mobile intranet cloud services, and robot control units.

Bio: Bill Huang is Founder and CEO of CloudMinds Inc. Before Cloudminds, he was GM of China Mobile Research Institute, SVP and CTO of UTStarcom, and previously worked at AT&T Bell Labs. Bill has creatively proposed the soft-switch concept of "the Network is the Switch", developed the first mobile soft-switch system in the world, and developed the first carrier-class streaming media exchange and IPTV system. He proposed the strategic concept of constructing the three major infrastructures (network, applications and terminals) of the next-generation mobile internet for the carriers, and promoted TD-LTE to be an internationally mainstream B3G standard, thus raising the technological influence of China in the communications industry.

In 2016, Bill was honoured IEEE CQR award. Besides, He is one of the first group of “Talent 1000” plan of China, a professor of University of Electronic Science and Technology of China, and a board member of GPS International Advisory Board of UC San Diego.

Bill received his master degree in Electrical Engineering and Computer Science from the University of Illinois in 1984. He graduated from the Huazhong University of Science and Technology in 1982 with a bachelor degree in Electrical Engineering.

Katja Mombaur, University of Heidelberg, Germany

Model-based Optimization for Humanoid and Wearable Robots​

In this talk, I give an overview of our research on motion optimization for humanoid and wearable robots. On the one hand, we are interested to improve the walking capabilities of humanoid robots in different terrains. Optimization based on realistic mechanical models of the robots is a very helpful tool since it can generate motions for such redundant, underactuated systems with multiple degrees of freedom and changing contacts that are feasible, stable and optimal. Optimization can also be applied to compliant robots.

On the other hand, we are interested to improve the design and control of wearable robots for the lower limbs and the lower back and other assistive devices. Using combined models of humans and the devices and optimal control, we can predict human movement in different conditions and determine the best possible support actions selecting passive and active components.

One important approach for both research fields is the solution of inverse optimal control problems based on recorded motion data which allows to identify objective functions underlying human movement. These optimality criteria can then be transferred to humanoid robots or be used for human movement prediction. For both fields, I will discuss the modeling levels to be used for describing humans and robots to address specific research questions. In addition, I will discuss possible combinations of optimal control methods with reinforcement learning and movement primitive approaches to reduce computation times and improve robot control.