Day 2 :
GunderFish LLC, USA
Time : 10:00-10:40
James P Gunderson has completed his PhD from the University of Virginia in 2003. His research has focused on Artificial Intelligence and the effective application of the same in automated systems and robotics. He has focused on Artificial Intelligence applications for over 30 years, and been part of two successful robotics/automation start-ups. He is the Growth Director for GunderFish LLC., which focuses on integrating cutting edge AI and predictive analytics with existing business flow. He has published more than 25 papers in journals and conferences.
The next wave of automation and robotics is benefiting from adding more \'intelligence\' to the systems deployed. The continuing decrease in cost and increase in processing power and sensor capability is leading a drive to smarter and smarter automated systems. However, there is a balance point between not smart enough and too smart.rnrnThis presentation will focus on three key elements:rn• What are the new capabilities and applications for the new smarter industrial and service robots?rn• What are the costs and potential problems introduced by these new artificial intelligence applications?rn• How does a systems engineer or designer determine how smart is smart enough?rnrnLike any other engineering decision, there are trade offs that lead toward an optimal design choice. Adding intelligence to an automated system is no different. However for many designers and engineers with the responsibility to make these critical decisions, there are no good \'best practices\' that can be applied. This often leads to decisions being made based on the input from the vendors alone, and this input is targeted towards the upper end of the decision space. I will present solid \'hands on\' tools for narrowing down the decision space to what is appropriate for the given problem. These tools will enable the responsible engineer to make solid \'best practices\' choices when looking at “How smart is smart enough?”
National University of Computer & Emerging Sciences, Pakistan
Time : 11:00-11:40
Asim ur Rehman Khan received BSc (EE) from UET, Lahore, Pakistan in 1981. He received MS (EE) from South Dakota State University, Brookings, South Dakota in 1987. He received PhD degree from Polytechnic University, now New York University, NY, USA in 1993. He worked in the Space Agency of Pakistan, SUPARCO, where he worked on the design & development of a small satellite. He has taught courses at SSUET, Karachi University, and NED University in Karachi, Pakistan. He worked in a software house for 5 years, where he was involved in the automation of MCI, USA fiber optic nation wide link. Since 2002, he is teaching at NU-FAST. His principle interests are image processing, nework protocols, and network security. He is a senior member of IEEE, and Pakistan Engineering Council (PEC).
The detection of important features of a moving object is a challenging task especially when these images are corrupted with heavy noise. This research proposes two statistical base techniques. The first technique performs three-way nested design using the analysis of variance (ANOVA). The three-way nested design corresponds to three-layers. The top layer is based on the temporal analysis where the model compares two consecutive image frames and identifies regions having sufficient temporal interframe changes. The next two layers perform statistical approach to see if there are sufficient intraframe variations. A large amount of intraframe variations are accounted for important features that may have edges to track across multiple image frames. In case of affirmative results in all the three layers, a second method based on the contrast function (CF) is used to identify edges in four possible directions. These four directions are horizontal, vertical, and two diagonal directions. The presence or absence of an effect is confirmed by testing a hypothesis. The test uses F-test, and Tukey’s T-test. The results are quite good for image frames that are previously corrupted with heavy Gaussain noise.
- Track 4: Robotics and Applications Track 6: Internet of ThingsTrack 7: Process and Energy Automation Track 8: Security in Manufacturing Industries
Tariq H. Tashtoush
Texas A&M International University, USA
Arizona State University, USA
Tariq Tashtoush is an Assistant Professor of Engineering in Texas A&M International University (TAMIU), Laredo, TX. He got his PhD and MS degrees in Systems and Industrial Engineering from State University of New York at Binghamton on 2013 and 2009, respectively and his BS in Electromechanical (Mechatronics) Engineering from Jordan University of Science and Technology (JUST), Irbid, Jordan in 2005. Currently, he is the faculty advisor and team leader of TAMIU Robotics and Intelligent Systems team and Students Engineering Council. In addition, he is the West Officer for the Corpus Christi Section of IEEE and the advisor and facilitator for the Youth Science Leader of Laredo non-profit organization. He is a multidiscipline engineer, who has experience in the field of Simulation and Systems Design, Production Quality and Management, Lean Manufacturing, Robotics and Automation, 3D Printing Processes, Engineering Statistical Analysis, Project Management, Optimization, Instruments and Electrical Devices, Reliability, Healthcare Systems, and Human Factors.
Human-Robot Interactions (HRIs) have been increasing in the last decades. This is because of a change in how items are being manufactured. Every time factories are looking for ways to optimize their current setting. This is to reduce costs and produce more by being efficient. One of the steps taken was to start diminishing the distance an item has to move from the beginning all the way to the final stage. In order to achieve shorter distances and smaller costs, humans and robots have started to work almost next to each other. This can lead to a more accident prone environment for the humans that work next to these robots. The main concern is that there are already more than 1 million robots in use worldwide. How can these become safer for the humans and in a cost effective way? This paper will explore the different technologies currently available to make the current robots human friendly and how the risk minimizing is done.
Hamid Marvi is an Assistant Professor of Mechanical and Aerospace Engineering at Arizona State University. Previously, he was a Postdoctoral Fellow at Carnegie Mellon University (2014-2015) and at Georgia Institute of Technology (2013-2014). He received his PhD in Mechanical Engineering from Georgia Tech in 2013. His work has been published in several scientific journals such as Science and has received attention from popular media such as New York Times, Los Angles Times, Washington Post, and BBC. His research aims to study fundamental physics behind interactions of biological systems with their surrounding solid, granular, and fluidic environments. Utilizing biological insights derived from these studies, he would like to develop bio-inspired robotic systems and programmable interfacial structures for search and rescue, exploratory, and medical applications.
Locomotion emerges from effective interactions with aerial, aquatic, or terrestrial environments. The majority of terrestrial terrain experienced by search-and-rescue or exploratory robots is flowing ground and is often composed of granular media. However, the lack of force models for granular environments has resulted in robotic systems that perform poorly on sandy hills. In contrast, many animal species and particularly snakes are highly versatile and remarkably successful at maneuvering on granular media. Thus, they can serve as sources of inspiration for transportation and robotic systems to traverse complex granular environments. In this talk, a series of experiments that help us understand the physics of sidewinder snakes’ interactions with a granular environment are presented. We found a control template for sidewinding on sandy inclines: The use of two orthogonal waves whose relative amplitudes are modulated is the key to successful climbing on sand. Next, the first snake robot capable of climbing sandy hills benefitting from this control template is demonstrated. Finally, my current research activities on developing structured active or passive elastomeric surfaces for crawling robots to help them interact more effectively with their terrestrial environments are discussed. Specifically, several active and passive mechanisms for the control of fibrillar friction and adhesion are presented. Using these mechanisms, we can readily control fibrillar friction in any planar direction and also take a major step toward the maximal adhesion of spatular elastomeric micro-fiber adhesives. The findings of these studies will result in the development of adaptive attachment structures and control methods for effective all-terrain search-and-rescue and exploratory robots.
Cranfield University, UK
Title: Understanding human robot collaboration in flexible and reconfigurable industrial environments
Time : 12:40-13:10
Phil Webb has a PhD in Manufacturing Engineering and is the Head of the Advanced Systems Centre at Cranfield University and the Deputy Director of the EPSRC Centre in Innovative Manufacturing in Intelligent Automation. He currently holds a Royal Academy of Engineering and Airbus Chair. During his career, he has worked with most of the major UK Aerospace companies including Airbus. Rolls-Royce, BAE Systems, Bombardier and GKN. He is an elected member of the Council of the British Robotics and Automation Association and has previously served as the UK representative to the International Federation for Robotics. His main research area is the implementation of flexible and reconfigurable automation in the aerospace industry with a strong focus on close human and industrial robot collaboration. He has published over 80 papers on robotics and automation.
Flexible and reconfigurable systems and human robot collaboration are key themes within current thinking on the wider introduction of robotics and automation in industry. However, it is important to understand the limitations of re-configurability and what the true value and use of human robot collaboration is. Most currently available collaborative systems rely on the use of force/torque limitation to ensure safety, which significantly limits the effectiveness of the systems. Also, since it is often the process (cutting, drilling, welding, etc.) that is the real source of danger, applications tend to be limited to simple handling tasks. This paper describes an approach to understanding how people can be more closely integrated within a reconfigurable manufacturing system using high payload industrial robots and multiple processes for the manufacture and assembly of complex high value products. The method analyses and breaks down human skill to understand how best to apportioned effort between people and robots and how to provide safety through the active reconfiguration of monitoring and safety systems and warning zones and also how to ensure that human operators feel comfortable working in such environments through the consideration of trust and acceptability. The resulting work is demonstrated and evaluated within a prototype human robot collaborative cell for the installation of flaps onto the wings of single aisle aircraft.
Chukwuemeka Odumegwu Ojukwu University, Nigeria
Title: Solving the problem of dearth of CAD software for consulting engineers in Nigeria and other developing countries: A case for software renting/subscription
Time : 13:55-14:30
Chidozie C Nwobi-Okoye obtained his PhD in Industrial and Systems Engineering from University of Benin, Nigeria. He is currently the Director of Postgraduate programs in the Faculty of Engineering of Anambra State University (Chukwuemeka Odumegwu Ojukwu University), Uli, Nigeria. He has published more than 32 papers in reputed journals and over 7 papers in local and international conference proceedings, and currently serving as a reviewer as well as in the Editorial Board of some reputable journals.
Finance is a major obstacle to software ownership in Nigeria and indeed most developing countries. This paper advocates the adoption of software rental/subscription as a way of ameliorating the problem of Computer Aided Design (CAD) software ownership among engineers in Nigeria and other developing countries. A survey of 50 engineers in Nigeria found that 100% agreed that software improves their designs, while 96% of them do not have access to all their software needs. 100% of those who do not have access to all their software needs, identified finance as responsible for their inability to access all their software needs, while 95.74% advocated software subscription/rental as a solution to financial obstacles to software ownership. This study therefore recommends the adoption of software rental/subscription to all CAD software vendors to improve software ownership among engineers in Nigeria and other developing countries. This will be a boost to the design and development of critical infrastructures needed for industrialization, growth and economic development.
Unified Computer Intelligence Corporation, Canada
Time : 14:30-15:00
Leor Grebler is Co-founder and CEO of Unified Computer Intelligence Corporation (UCIC), a company dedicated to bring voice interaction to hardware. Its initial product - Ubi – The Ubiquitous Computer – was a voice activated computing device that offered instant access to information and control of home automation devices and was the first product to offer natural environment-based voice interaction. He steers UCIC towards its goal of making interaction with technology more human and natural.
With far field technologies finally getting to market-ready pricing, we are about to see a large proliferation of environment based voice interactive products. Soon, even through noise and multiple speakers, we'll be able to interact in our home through appliances and specialized devices and in office and retail settings through voice interaction. However, there still exists challenges to build voice interaction. While artificial intelligence is able to predict users desires and answer complex queries, it still requires correct inputs from the user and an ability to effectively communicate results of requests. This talk will review the current state of voice interactive technologies, the areas of development, and discuss where advances will affect human computer interaction in the next 5-10 years.
Methodist University of Piracicaba, Brazil
Title: Analysis of the integration between information tools discrete event simulation and production system
Time : 15:00-15:30
Rodrigo Ferro has obtained his Graduate degree in Industrial Mechanical Engineering from the Methodist University of Piracicaba (2006) and Master’s degree in Production Engineering from the Methodist University of Piracicaba (2014). He is currently a Professor of Production Engineering at the Methodist University of Piracicaba and Researcher Level D at CNPq.
The discrete event simulation is a tool that has been used in production systems to aid decision making in order to increase speed and assertiveness of decisions. However, the modeling phase of discrete event simulation becomes a limitation in a decision making process that requires the response speed. This happens due to the time of collection of information from the production system and the time required for the processing of the information collected. This study aims to analyze the information integration tools that assist in the flow of data between production systems with discrete event simulation models allowing streamline the modeling stage and analysis of data generated in the simulation. Initially an exploratory research in scientific literature on the main tools used in managing and analyzing information between discrete event simulation and production systems was carried out. Then it was possible to develop a map with the integration of information between the production system and discrete event simulation both on-line form when off-line. Analyzing the map developed is possible to identify the points to be developed to move forward in an online simulation system and integrated manufacturing.
Texas A&M International University, USA
Time : 15:30-15:50
Patricio Rodriguez is currently pursuing a degree in Systems Engineering at Texas A&M International University. He has been a part of Engineering Organization Society of Hispanic Professional Engineers (SHPE). He was able to be the regional student representative for region V (TX, OK, AR, LA, MS, AL, TN) followed by president for the chapter at Texas A&M International University. His future plans are to pursue Master’s degree in Industrial Engineering and a PhD in the same
Shipping and handling costs account for 25-75% of the total cost of a product. Laredo, Texas is known as biggest land port in the south border of the United States. The economy in this area is dominated by the logistics and warehousing business sector. The logistics sector has slowly started to integrate automated processes within the warehousing departments. Clients demand lower prices thus affecting the economy of the business sector. Automating the warehouses can be a costly project that slowly has started to be done. Autonomous forklifts have started to be tested. Safety technology will be analyzed. This paper will focus on studying the advantages and disadvantages of using autonomous forklifts and the impact that this change in technology can bring. This can be a call for ways to make the automatization process more affordable as new technology emerges.
University of Regina, Canada
Title: Design and development of a learning-based human-analogous real-time control method with applications in industrial automation
Time : 15:50-16:10
In the animal kingdom, humans learn by acting on their environment, observing the consequence or effect of their acts, and learning to adjust their actions accordingly over time to improve the output generated by their actions. They also learn to optimize their actions through reinforced learning. The ability of humans to optimize their behavior in natural systems can be expanded to that in man-made engineered systems as well. There is a world-wide effort towards developing control systems that are inspired from human intellect. This research is a timely response to this global need. The focus of this research is to develop a learning-based human-analogous control method that can be incorporated into the process of industrial automation in a short time and in a systematic way. This research addresses the design of Human Machine Interfaces (HMI) for re-enforced learning that can replace the time-consuming process of model-based control strategies currently exercised via the from-the-first-principle modeling paradigm. The real-time I/O data obtained within a Human-In-The-Loop (HITL) control system are used to: (1) optimize the design of the HMI, (2) optimize the learning curve associated with the trials via a consistency matrix, and (3) design optimal unmanned control strategies based on the most-consistent I/O data obtained via the real-time experiments in the aforementioned HITL simulator. Case studies are provided for a benchmark control problem, namely the servo control of a rotary actuator with unknown dynamics.