Plenary Talk 1: Dr. Antonio Moran
President IEEE Robotics and Automation Society - Peru Section
Dr. Antonio Moran obtained the doctor and master degrees from Tokyo University of Agriculture and Technology, Japan, where also worked as professor and research scientist in the Department of Mechanical Systems Engineering. Dr. Moran has been Director at Peruvian University of Applied Sciences and professor at graduate schools in Peruvian universities. Dr. Moran pioneered the introduction of artificial intelligence concepts in Peru as well as the development and applications of neural networks, fuzzy logic and genetic algorithms for designing intelligent control systems. Dr. Moran is visiting professor at Stockholm University, Sweden, and at present is president of the IEEE Robotics and Automation Society, Peru Section. His research interests include intelligent systems design and applications of fuzzy-neural networks to engineering, medicine, finance, operations research and other fields.
Plenary Talk : ”Intelligence Technology for Thinking Robots”
Trying to conciliate several approaches, human intelligence can be defined as the mental capability for reasoning, comprehending, learning, planning, identifying relationships for acting and solving problems to improve our living state. In a simple way, it can be said that intelligence is the capability for comprehending the surroundings to figure out what to do and act accordingly.
The ability to think has been one of the most fascinating features engineers have been trying to incorporate into robots since the very beginning of the robotic concept and along its evolution: initial industrial robots, service robots, ubiquitous robots, genetic robots and bio robots have incorporated increased levels of intelligence reflected in their autonomy and their modular, versatile and ubiquitous capabilities.
Humans attempt to solve problems by using intelligence and knowledge based on data and information. However, robots require intelligence technology to solve the problem because they do not possess the knowledge to process data and information properly. Similarly as humans, robot intelligence should integrate the different spheres of decision-making capabilities which include cognitive intelligence, social intelligence, behavioral intelligence, ambient intelligence, collective intelligence and genetic intelligence. The ultimate goal is to integrate all these categories of intelligence into a humanoid robot working in home and office environments in a friendly and cooperative way for making our lives more comfortable and productive.
Plenary Talk 2: Dr. Manfred Huber
Manfred Huber is an Associate Professor in the Department of Computer Science and Engineering at the
University of Texas at Arlington. He received his PhD in Computer Science at the University of Massachusetts, Amherst in 2000. He is the director of the Learning and Autonomous Robot Laboratory (LEARNLab) and a co-director of the SmartCare Laboratory and focuses on research in Cognitive Development, Autonomous Learning Robots and Machine Learning techniques for Health Monitoring and Intelligent Environments
Plenary Talk : ”Developmental and Reinforcement Learning for Autonomous Robots”
Autonomous robot systems have advanced significantly over the last decades, enabling them to increasingly operate in less structured environments and less predictable task domains. In recent years, a number of specialized robotic systems have entered domains such as driving in regular traffic, assisting during surgeries, performing limited autonomous cleaning tasks in the home, or performing assistive functions with humans.
While most of these are still demonstration systems, they are showing capabilities that make it possible for them to operate in close to real-world domains. While this bodes well for the advent of service and assistive robots in general, real-world environments, the competency and applicability of most of the current systems are still limited to a relatively small niche in terms of tasks and parts of the world they can correctly represent and reason about.
Biological systems, in contrast, have the ability to adapt to incrementally larger task domains and environmental conditions by learning to perform new task as well as by forming new representations that allow them to successfully encapsulate the complexity of the world and that provide them with compact means to reason about novel situations. To permit autonomous robot systems to show similar capabilities without the need for an always present human teacher, Reinforcement Learning techniques have been increasingly applied to robot systems and combined with developmental learning approaches which form a basis for life-long learning and allow the system to incrementally learn more complex tasks and representations while reducing the effect of the curse of dimensionality. This talk will present and discuss some examples of developmental learning systems using Reinforcement Learning applied to autonomous robot tasks. It will introduce basic developmental principles that, in combination with Reinforcement Learning techniques, can be applied to learning robot systems in order to to allow them to develop skills and representations on-line and re-use these along a developmental trajectories in order to simplify subsequent learning problems.
Plenary Talk 3: Dr. Teodiano Freire Bastos
Dr. Teodiano Freire Bastos received his B.Sc. degree in Electrical Engineering from Universidade Federal do Espirito Santo (Vitoria, Brazil) in 1987, his Specialist in Automation degree from Instituto de Automatica Industrial (Madrid, Spain) in 1989, and his Ph.D. degree in Physical Science (Electricity and Electronics) from Universidad Complutense de Madrid (Spain) in 1994. He made two postdocs, one at the University of Alcalá (Spain, 2005) and another at RMIT University (Australia, 2012).
He is currently an Associate Professor in Universidade Federal do Espirito Santo (Vitoria, Brazil), teaching and doing research at the Postgraduate Program of Electrical Enginneering, Postgraduate Program of Biotechnology and RENORBIO Ph.D. Program. His current research interests are signal processing, rehabilitation robotics and assistive technology for people with disabilities.
Plenary Talk : “Dispositivos Robóticos de Asistencia para Personas con Discapacidad”
The goal of this speech is to show some developments in Rehabilitation Robotics carried out in UFES/Brazil. First, a robotic wheelchair will be introduced. This wheelchair can navigate in autonomous mode, or auto-guided (following metallic strips), or commanded by:
- Eye Blinks
- Eye Movements
- Head Movements (Using Accelerometer or Vision)
- Brain Signals
The Robotic Wheelchair has a laser sensor to detect obstacles and contains also a communication system onboard, which can be command through the aforementioned signals.
The second development to be introduced is the robotic walker which uses force sensors to detect the motor intention and laser sensor to measure the human gait. It can be used inside rehabilitation clinics and help the recovery of people with problems associated with mobility. The robotic walker is a tool for also helping the diagnostic of diseases such as osteoarthritis and can be used for analysis and study of the human gait.
Plenary Talk 4: Dra Luciane dos Santos
Holds a degree in Education from the State University of Rio de Janeiro, Masters in Education from the University of Rio Grande do Norte and Ph.D. in Education from the University of Rio Grande do Norte. He is currently Professor at the Federal University of Rio Grande do Norte. Has experience in the area of education, with an emphasis on educational policy, mainly in the following areas: educational management, organizational culture, political-pedagogical project and institutional assessment, Educational Robotics. Participate in various research projects in the areas of interest.
Plenary Talk: “Robótica educacional como estratégia de ensino: uma discussão contextualizada”
Plenary Talk 5: Dr. Luiz Marcos Goncalves
Luiz Goncalves holds a Doctorate in Systems and Computing Engineering from Federal University of Rio de Janeiro, graduated in 1999. He is Associate Professor at the Computing Engineering Department of Federal University of Rio Grande do Norte, Brazil.
He has edited 3 books (proceedings), published more than 170 papers, including journals, magazine, and conference proceedings. He is advisor of several graduate (more than 30 formers, master and doctorate degrees) and also undergraduate students (more than 60 formers, including supervision of undergraduate research work and final course work). Has participated in many international events and served in the Program Committee of many conferences, including contributions as Program Chair, General Organizing Chair, PC Member and Reviewer. He has also being Associate Editor of Journal JINT. He has done researches in the several aspects of Graphics Processing including fields as Robotics Vision (main interest), Computer Graphics, GIS, Geometric Modelling, Animation, Image Processing, Computer Vision, and Digital Television. He has coordinated and participated in many projects in the above fields. He was the Chair of Brazilian Computer Graphics and Image Processing Commitee and of the Robotics Commitee, both under Brazilian Computer Society. He was the chair for Robocup Latin American Committee (2005-2009) and he is member of the IEEE Latin American Robotics Council (since 2002).
Plenary Talk: “Robotica Educacional, experiencias en Brasil”
Plenary Talk 6: Ikuo Mizuuchi
Ikuo MIZUUCHI has been an Associate Professor at the Division of Advanced Mechanical Systems Engineering at Tokyo University of Agriculture and Technology since 2009. He received B.E. in Mechanical Engineering from Waseda University in 1995, M.Eng. and Ph.D. both in Mechano-Informatics from The University of Tokyo in 1998 and 2002, respectively. He was appointed as a Research Fellow of Japan Society of the Promotion of Science in 2000, a Project Assistant Professor in the Graduate School of Information Science and Technology at the University of Tokyo in 2002, and a Senior Assistant Professor in the Department of Mechano-Informatics at the University of Tokyo in 2006.
Plenary Talk: “Musculoskeletal Humanoids: Human-Inspired Design of Hardware and Software”
Human’s body gives various inspirations in thinking of future robotics. I and my colleagues have proposed the ‘musculoskeletalhumanoid’ approach, and have been developing a series of musculoskeletal humanoids. The potential advantages of human-inspired design are utilization of mechanical elasticity, adaptivity to diverse tasks and environments, autonomous development of motions and behaviors, and so on. I will introduce my studies on design, implementation, control, learning, etc. of musculoskeletal humanoids, and recent works on utilizing natural elasticity for explosive motions.