Workshops
The Robotics conference is pleased to present an innovative series of workshops on current research topics in robotics. The following describes these workshops in detail.Workshops will occur on Saturday, June 11th
- Workshop on Surgical Simulation based on Reality-Based Soft-Tissue Models
- Workshop on Humanoid Manipulation
- Workshop on Learning for Locomotion
- Self-Reconfigurable Robots/Systems and Applications Workshop
- Workshop on Modular Foundations for Control and Perception
- Robotics Education Workshop
Workshop on Surgical Simulation based on Reality-Based Soft-Tissue Models
URL: prism.mem.drexel.edu/desai/SurgicalSimulation_Workshop/
This workshop will focus on measurement and modeling of various types of tissue interaction relevant to surgical training and simulation (such as probing, cutting, dissecting, percutaneous, etc.). Research in measurement of tissue properties for the development of improved surgical simulators and robot-assisted surgery has blossomed in recent years. Advances in computer graphics have enabled realistic interactive simulation of very large deformable models, due to dramatic changes in commodity graphics hardware and the development of novel simulation algorithms. The workshop brings together leading researchers in these two fields. The presentations will provide an overview of research results needed to develop improved systems for surgical simulation and robot-assisted surgery. One of the main goals of this workshop is to foster discussion among participants, to identify challenging issues, and to form a consensus on: 1) the primary barriers to developing accurate tissue models, 2) how tissue models should be structured for application in robot-assisted surgery and surgical simulation, and 3) what are the unique requirements of tissue models that should guide future algorithm development.
Workshop on Humanoid Manipulation
URL: groups.csail.mit.edu/lbr/event/rss05/index.shtml
This workshop is a forum to foster discussion among researchers about the emerging field of humanoid manipulation in which robots safely coexist with humans and usefully manipulate objects found in unstructured and built-for-human environments. The workshop will focus on the fundamental research questions facing the field, including:
- Perception and manipulation of the unfamiliar.
- Achievement of human level dexterity and sensitivity.
- Biological foundations in human manipulation.
- Mobile manipulation in built-for-human environments
- Development of advanced perceptual systems for unstructured environments
- Minimal-model approaches to grasping
- Novel solutions in sensing and actuation
- Mechanisms for human-like dexterity
- Development of compliant and force controlled manipulators
- Computational and neural models of the human motor system
- Advances in high density tactile sensing
- Development, learning, and adaptation for reaching and grasping
- Cognitive architectures for manipulation
- Learning of object affordances
- Behavior-based approaches
- Review of the state-of-the art
- Manipulation as a social, interactive engagement
- Bimanual and full-body grasping
- Applications of humanoid manipulation
The workshop will include invited presentations, panel discussions, a video and poster session, and live humanoid demonstrations. Please visit our website for details on submission and participation.
Workshop on Learning for Locomotion
URLS:
www-clmc.usc.edu/~jrpeters/workshop.html
www.jan-peters.net/Research/LearningForLocomotion
Over the last few decades, there has been an impressive amount of published work on legged locomotion, including bipedal walking, running, hopping, stand-ups, summersaults and much more. However, despite all this progress, legged locomotion research has largely been driven by researchers using human insight and creativity in order to generate locomotion control algorithms. In order to improve the robustness, energy efficiency, and natural appearance of legged locomotion, there may be a significant advantage to using machine learning methods to synthesize new controllers and to avoid tedious parameter tuning. For instance, it could be advantageous to learn dynamics models, kinematic models, impact models, for model-based control techniques. Imitation learning could be employed for the teaching of gaits patterns, and reinforcement learning could help tuning parameters of the control policies in order to improve the performance with respect to given cost functions. In this context, we would like to bring together researchers from both the legged locomotion and machine learning in order to discuss which locomotion problems require learning, and to identify the machine learning methods that can be used to solve them.
Self-Reconfigurable Robots/Systems and Applications Workshop
URL:http://www.isi.edu/robots/workshop2005/
Self-reconfigurable modular robots are metamorphic systems that can autonomously change their logical or physical configurations (such as shapes, sizes, or formations), as well as their locomotion and manipulation, based on the mission and the environment in hand. Because of their modularity, versatility, self-healing ability and low cost reproducibility, such robots provide a flexible approach for achieving complex tasks in unstructured and dynamic environments. They are well suited for applications such as search and rescue, reconnaissance, self-assembly, inspections in hazardous environments, and exploration in space and ocean. They also pose fundamental research challenges for robotics and other major branches of computer science, mechatronics and control theory.
The challenges are due to the dynamic topology of the network of modules, the limited resource (power, size, torque, precisions, etc.) of individual modules, the difficulties in global synchronization, the preclusion of centralized decision makers, and the unreliability of communication among modules. This workshop will present the recent progress in the research community for these challenging tasks and their real-world applications in space and other related fields. We will present distributed control architecture and algorithms, discuss the ability of plug-and-play mechatronics parts and arbitrarily reshuffling modules (body-parts) in systems, discuss the recent theoretical development for self-reconfigurable systems, analyze the hardware/software challenges we face to make these robots for multifunctional applications, and outlook the future of this exciting research topic.
Workshop on Modular Foundations for Control and Perception URL: http://www.cse.unr.edu/~monica/Conferences/rss_wmfcp05.html
Advances in the field of robotics have begun to realize
sophisticated robotic platforms with increasingly richer sensing
requirements. In addition to increasing robot functionality, robotics
researchers also strive towards increasing the accessibility of robot
control to greater segments of society. The balancing of greater robot
functionality versus greater robot accessibility can be difficult. In
addressing these issues, greater emphasis has been placed towards creating
and building upon basic modular components. Using neuroscience and biology
as inspiration, such modules reduce of the complexity of the control and
sensing interfaces while providing necessary functionality.
More specifically, we consider a module (or "primitive" or "behavior") to
encodes a link between perception and motor control. This link furthers the
computational process that enables a robot to achieve or maintain
certain goals. A basis set of such primitives is sufficient, through various
combination operators, for generating the entire movement/activity
repertoire of a robot. As a consequence the development, learning and
utilization of such primitives robot control has become an area of high
interest in high DOF systems (such as humanoids), learning by imitation or
demonstration and understanding of user activity or intent.
The workshop will explore modular foundations for robot control and
perception and will address three basic areas:
1) development and learning of modular primitives
2) utilization of modular foundations for both control and perception
3) utilization of known primitives for symbol grounding in higher level
methods
Robotics Education Workshop
URL: http://projects.csail.mit.edu/rss/RobotEd/
Robotics provides the perfect educational tool for introducing students to embedded systems and computation for interacting with the physical world. Several universities have already introduced special topics courses on robotics. The curricula and hardware platforms for these courses is quite varied and the goals of the courses differ across ME, EE, and CS departments. We wish to leverage this excellent body of knowledge and discuss how to develop an integrated approach to teaching robotics that will train students simultaneously in in foundational aspects of designing, controlling, and programming robots and embedded systems. The objective of this workshop is to evaluate the state of the art for undergraduate robotics education and discuss how to build on this experience toward a broad integration of robotics in the undergraduate curricula.
Our proposed workshop will cover topics including: robotics curricula, hardware platforms and kits, software platforms, issues related to integrating EE, ME, and CS topics in one course, laboratory curricula, and role of broad challenges and competitions in the curricula.