COLLOQUIUM Department of Computer Science and Engineering University of South Carolina Lantao Liu Department of Computer Science and Engineering Texas A&M University Date: April 11, 2013 Time: 1400-1500 (2:00pm-3:00pm) Place: Swearingen 1A03 (Faculty Lounge) Abstract Like humans, multiple robots need to collaborate with each other to accomplish certain missions efficiently. Task allocation (or assignment) is an important component for building multi-robot systems that behave as a coordinated whole. This talk will cover two settings in which coordination methods using task allocation techniques are effective; the two settings bring together ideas from different fields: distributed AI, operations research, graph theory, high performance computation, and economics. Specifically, I will introduce two decentralized task assignment methods: (1) a multi-level assignment decentralization framework based on matrix partitioning in order to cluster the strongly related robot-tasks pairs, and (2) a task-swap based assignment algorithm with many unique features that are particularly suitable for distributed multi-robot task allocation. Furthermore, I am also going to discuss how the classic assignment algorithms can still be extended to solve new problems in the multi-robot domain, e.g., how the combinatorial assignment algorithm itself can be extended to tackle the task allocation involving uncertainty; and how the matching bigraph data structure of the assignment algorithm can be manipulated to solve the robotics routing and formation control problems. Lantao Liu is a Ph.D. candidate (with dissertation successfully defended) in the Department of Computer Science and Engineering at Texas A&M University. Liu works on developing coordination methods for multiple robots. His dissertation was on task allocation problems. Liu also works on estimating and analyzing the properties of swarm systems, motion planning for robotics, and protein folding in computational biology. Recently, Liu has been awarded the prestigious Texas A&M University Dissertation Fellowship and the Chinese Government Scholarship for outstanding students abroad, which is based on nominations by Chinese embassies worldwide. Liu graduated with highest honors (academic silver medalist) from the Department of Automatic Control at Beijing Institute of Technology in 2007.

COLLOQUIUM Department of Computer Science and Engineering University of South Carolina Jingjin Yu Department of Electrical and Computer Engineering University of Illinois at Urbana Champaign Date: April 4, 2013 Time: 1330-1430 (1:30pm-2:30pm) Place: Swearingen 2D05 (Former Staff Lounge) Abstract As we know, Kalman filters, given their effectiveness in handling probabilistic events, play an important role in the processing of probabilistic sensor information, such as point clouds gathered from laser scanners. However, not all data is amenable to probabilistic analysis - some problems induce inherently combinatorial data structures. For these problems, a combinatorial filtering approach is necessary. In this talk, I discuss two such robotics (sensor fusion) problems. The first problem generalizes visibility-based pursuit-evasion games and seeks to maintain the distribution of hidden targets that move outside the field-of-view of the pursuers while a sensor sweep is being performed. Observing that key events happen only when the shadow region (all points invisible to the sensors) changes combinatorially or when targets pass in and out of the field-of-view, we can significantly reduce the amount of sensor information that we store, without discarding any useful information for computing the distribution of the hidden targets. This allows us to "live" inside a erived information space, which gives rises to efficient algorithms. These filtering algorithms provide critical information for tasks such as counting, herding, pursuit-evasion, and situational awareness in general. Next, we study the problem of using sparse, heterogeneous sensors to verify the stories (i.e., path samples) of agents (robots or people). Since there are two sets of data, the combinatorial filter for this problem can be built in two ways: Using a filter (an automaton) built from sensor data to process the story or using a filter built from the story to process the sensor data. Both approaches lead to dynamic programming based (optimally) efficient algorithms. In addition to exact path inference, our method also applies to approximate path inference that allows errors in the story. Besides immediate applicability toward security and forensics problems, the idea of external verification is promising in complementing design time model verification. Jingjin Yu received a B.S. in materials science and engineering from USTC, Hefei, China (1998). He holds M.S. degrees in chemistry (University of Chicago, 2000), mathematics (University of Illinois at Chicago, 2001), and computer science (University of Illinois at Urbana Champain, 2010). He is currently a Ph.D. candidate with the Department of Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign. His research interests include robotics and control theory.

Sidney Valentine ABSTRACT Supervisory Control and Data Acquisition (SCADA) systems are widely used in automated manufacturing and in all areas of our nation's infrastructure. Applications range from chemical processes and water treatment facilities to oil and gas production and electric power generation and distribution. Current research on SCADA system security focuses on the primary SCADA components and targets network centric attacks. Security risks via attacks against the peripheral devices such as the Programmable Logic Controllers (PLCs) have not been sufficiently addressed. Our research results address the need to develop PLC applications that are correct, safe and secure. This research provides an analysis of software safety and security threats. We develop countermeasures that are compatible with the existing PLC technologies. We study both intentional and unintentional software errors and propose methods to prevent them. The main contributions of this dissertation are: Develop a taxonomy of software errors and attacks in ladder logic Model ladder logic vulnerabilities Develop security design patterns to avoid software vulnerabilities and incorrect practices Implement a proof of concept static analysis tool which detects the vulnerabilities in the PLC code and recommend corresponding design patterns.

COLLOQUIUM Ioannis Rekleitis School of Computer Science McGill University Date: March 28, 2013 Time: 2:00pm-3:30pm Place: Faculty Lounge, Swearingen 1A03 Abstract The last few years, robots have moved from the pages of science fiction books into our everyday reality. Currently, robots are used in scientific exploration, manufacturing, entertainment, and household maintenance. While the above advances were made possible by recent improvements in sensors, actuators, and computing elements, the research of today is focused on the computational aspects of robotics. In particular, methodologies for utilizing the vast volumes of data that can be generated by a robotic mission, together with techniques that would allow a robot to respond adequately in unforeseeable circumstances are the challenges of tomorrow. This talk presents an overview of algorithmic problems related to robotics, with the particular focus on increasing the autonomy of robotic systems in challenging environments. Cooperative Localization Mapping and Exploration employs teams of robots in order to construct accurate representations of the environment and of the robot's pose. The problem of coverage has found applications ranging from vacuum cleaning to humanitarian mine removal. A family of algorithms will be presented that solve the coverage problem efficiently in terms of distance travelled. Interestingly, the planetary and underwater exploration share many challenges when viewed from an algorithmic point of view. An a priori unknown environment and limited communications are among the most obvious. I would present recent results from a multi-robot experiment employing a UAV, a USV, and a AUV operating in sync with a remote marine biologist located thousands of kilometers away. Ioannis Rekleitis is an Adjunct Professor at the School of Computer Science, McGill University, Montreal, Canada. Between 2004 and 2007 he was a visiting fellow at the Canadian Space Agency. During 2004 he was at McGill University as a Research Associate in the Centre for Intelligent Machines with Professor Gregory Dudek in the Mobile Robotics Lab (MRL). Between 2002 and 2003, he was a Postdoctoral Fellow at the Carnegie Mellon University in the Sensor Based Planning Lab with Professor Howie Choset. He was granted his Ph.D. from the School of Computer Science, McGill University in 2002 under the supervision of Professors Gregory Dudek and Evangelos Milios. Thesis title: "Cooperative Localization and Multi-Robot Exploration." His Research has focused on mobile robotics and in particular in the area of cooperating intelligent agents with application to multi-robot cooperative localization, mapping, exploration and coverage. His interests extend to computer vision and sensor networks. He has worked with underwater, terrestrial, aerial, and space robots. Ioannis Rekleitis has published more than sixty journal and conference papers. His work can be found online at: http://www.cim.mcgill.ca/~yiannis/.

by Dr. Duncan Buell Scholars in the humanities who are moving into the digital world are confronting a complex issue: how to balance work on and use of digital tools with “research” in their respective disciplines. Are some scholarly activities real research, while others are “just work”? This is an issue that has been faced in the sciences, where it has been resolved at least in part and relates to the serious issue of “credit” for the creation of digital artifacts. While you eat your lunch, Buell will give his view on how to distinguish work and research, as well as the role of making tools in support of and as part of scholarly disciplinary work. Coffee will be provided. More information here.
Video streaming by Ustream

Go to POSSCON.org for more details.

COLLOQUIUM Department of Computer Science and Engineering University of South Carolina Jenay M. Beer School of Psychology Georgia Institute of Technology Date: March 20, 2013 Time: 1430-1530 (2:30pm-3:30pm) Place: Swearingen 2A31 Abstract Many older adults wish to remain in their own homes as they age. However, challenges threaten an older adult’s ability to age in place, and even healthy independently living older adults experience challenges in maintaining their home. Challenges with aging in place can be compensated through technology, such as home assistive robots. However, for home robots to be adopted by older adult users they must be designed to meet older adults’ needs for assistance and the older users must be amenable to robot assistance for those needs. In this talk, I will discuss a range of projects (both quantitative and qualitative in nature) assessing older adults’ social interpretation, attitudes, and acceptance of assistive robotics. Study findings suggest that older adults’ assistance preferences discriminated between tasks, and the data suggest insights as to why older adults hold such preferences. The talk will detail a multidisciplinary approach to studying human-robot interaction (HRI) and how findings from user studies can apply to preliminary design recommendations for future assistive robots to support aging in place. Jenay M. Beer is a 6th-year Ph.D. student in Engineering Psychology at Georgia Tech. She is a member of the Human Factors and Aging Laboratory, co-directed by Wendy A. Rogers and Arthur D. Fisk. Her research intersects the fields of Human Robot Interaction (HRI) and Psychology. Specifically, she studies home-based robots designed to assist older adults to maintain their independence and age in place. She has studied a variety of robotic systems and topics such as emotion expression of agents, user acceptance of robots, healthcare robotics, and the role of robot autonomy in HRI. Jenay received a B.A. degree in Psychology from the University of Dayton, Ohio and an M.S. in Engineering Psychology from Georgia Tech.

COLLOQUIUM Department of Computer Science and Engineering University of South Carolina Cory Henson Kno.e.sis Ohio Center of Excellence in Knowledge-enabled Computing Wright State University Date: March 7, 2013 Time: 1430-1530 (2:30pm-3:30pm) Place: Swearingen 1A03 (Faculty Lounge) Abstract There are an estimated 40+ billion sensors connected to mobile devices; and it has been predicted that within the next five years, sensor data from such devices will become the dominant type of information on the Web. For this data to be useful in applications, ranging from healthcare to environmental monitoring, it needs to be translated into higher-levels of abstraction (e.g., translated from observed symptoms to disorders). This act of translating low-level signals into high-level knowledge is called perception, and while people have evolved sophisticated mechanisms to efficiently perceive their environment, machines continue to struggle with the task. In this talk, I will describe a model of machine perception, Intellego, derived from cognitive theories of perception. Encoded in the Web Ontology Language (OWL), this model provides a formal semantics of perception by defining the information processes involved in collecting and interpreting heterogeneous sensor data. While the use of OWL enables advanced integration and interpretation of sensor data, the computational complexity of reasoning seriously limits its applicability and use within resource-constrained environments, such as mobile devices. To overcome this challenge, I will discuss an efficient implementation using bit-vector encodings and operations, resulting in order-of-magnitude improvements in both efficiency and scale. The applicability of this approach will be demonstrated through a real-world application in healthcare. This application, called kHealth (knowledge-enabled healthcare), provides a means for patients with chronic heart disease, and their doctors, to remotely monitor and manage their condition after release from the hospital. The kHealth application is expected to go into pre-clinical trials by the end of March. Cory Henson is a Ph.D. Candidate at Wright State University and a researcher at Kno.e.sis, the Ohio Center of Excellence in Knowledge-enabled computing. He also works at Riverside Research, a research organization spun out of Columbia University, where he collaborates on projects for the DoD and Intelligence Community. He has bachelor's degrees in both Computer Science and Cognitive Science from the University of Georgia. Cory's research revolves around Web-based data and metadata management, including knowledge representation, ontology modeling, data integration, and reasoning. More specifically, his research comprises the synergistic use of semantic web, artificial intelligence, and cognitive science to develop the theoretical foundation and practical tools needed to enable sensor data management and machine perception. This technology has exciting applications in developing smart environments and enabling more proactive, preventative healthcare. For more, see: http://knoesis.org/researchers/cory.